Realign code with the 2026-07-31 findings-resolution rulings

The full bullet list from Implementation card bf080d9a — both ruling
batches, including the three appended mid-session by 16ef377:

- Restore subjects compose the inverse, never nest: crossing "Undo: S"
  emits "Redo: S" and vice versa; parity, not stack depth, reads a
  legacy double prefix (GitHistoryProvider.restoreSubject).
- Git-operation failures join the one-shot failure banner tier:
  BannerCenter.GitFailureBanner (undo/redo/branchSwitch/addGit), error
  tone at failure rank merged with write one-shots by recency; the
  postLoss compromise is retired at both AppModel wirings.
- order/schema optional below the board root: append-at-end reading
  (ordered siblings first, folder-name tie-break among the order-less),
  schema reads 1, both coerce-tier logged; the root keeps its
  requirements. Ranks.resolvedOrders materializes finite ranks so
  models and placement math stay untouched; first Writer rewrite
  stamps a real rank on touch, placement against an order-less sibling
  stamps that sibling inline in the same bracket. Agent guide v10
  teaches optional keys and zero-read filing. Hostile-YAML order
  shapes become coercion tests; Fixtures/Valid/optional-keys.kanban
  replaces the four retired Malformed boards.
- .gitignore is the relocation-heal noise gate: GitignoreRules pure
  matcher (standard semantics, board-root file only), loader consults
  it once per walk so matched loose files keep the stray posture;
  seeded (.DS_Store + .*.lanework-*) at board creation and template
  instantiation, healed in when missing at open — repo-nested
  included; empty file honored, existing files never edited; the
  committer's obedience via libgit2 status is pinned by test.
- Comments crash-residue sweep gates on step ownership: HistoryStep
  derives backing from its own undo expectations, backedContent unions
  both stacks, the sweep purges per-entry only what no live step owns.
- Skip-purge decoupled (16ef377): a stale-skipped coarse step strands
  whole in NativeHistoryProvider.strandedSteps — still backing, retired
  only at session end; clean exits purge as before.
- Coarse close step named "Changes to '<card>'"; the fine body-edit
  wording never leaks onto the board menu.
- Branch-switch settle clears every open card window's fine stack on
  Save All and Discard alike; the empty fold registers no coarse step.
- Close flush awaits its covering snapshot (quiesce + one generation
  bump, 1s bound), and an explicit flush now queues behind an
  in-flight one instead of skipping — the audit-caught interleaving
  could lose a close flush permanently when the debounce fired inside
  the close sequence; regression tests force both races.
- Commit comment bullets sort chronologically by created, not UUID.
- The production-unwired CardBodyEditSession.editSessionDidChange seam
  is deleted with its seam-only tests.
- Composition-root pins: beginSession composes the committer with the
  store's own EchoLedger and binds the announcer (the miswire class).
- Deliberate 06 conformance pass over every 2026-07-31-tagged
  sentence: fixed Change-custom-key subjects (the retired named
  generic was the only producer), the unbuilt Replace attachment
  vocabulary, heal commits now authored Lanework Integrity, the config
  reader scopes identity to plain [user] sections, add-git re-runs
  detection at create (a stale mode-none could initialize inside the
  user's repo), and add-git failures answer at the form or the banner.
  Structural residue filed on the Redesign board.

2554 tests / 439 suites green.

Claude-Session: https://claude.ai/code/session_01CqjXB7ASoWtbyoGod68k97

Kanban/UI/Board/BoardGitControls.swift

@@ -235,8 +235,15 @@ struct BoardGitControls: View {
 // MARK: - Commit identity
 
 /// **The name and email that repo-local `.git/config` carries** (06-history-undo.md ▸ Interaction
-/// with external writers: "The board popover's git section exposes name/email fields that write that
-/// repo-local config — the setting *is* the file, portable to any git client, per-board by nature").
+/// with external writers: "The board settings sheet's identity section … exposes name/email fields
+/// that write that repo-local config — the setting *is* the file, portable to any git client,
+/// per-board by nature").
+///
+/// **Its home is the sheet, and the sheet is not built yet.** The 2026-07-31 popover/sheet split moved
+/// every setup-shaped control — add-git, branch creation, these fields, remote and credentials — out
+/// of the popover and into a board settings sheet (03-board-ui.md ▸ Board settings sheet), leaving the
+/// popover the daily face. This view is the fields, hosted where they were; the sheet's card moves
+/// them, along with the visibility-scoped poll below, which 06 says "rides with the fields".
 ///
 /// ### The placeholder is the whole of the identity rule made visible
 ///
@@ -270,8 +277,11 @@ private struct BoardGitIdentityFields: View {
         case email
     }
 
-    /// How often an open popover re-reads the config file. Slow enough to be free, fast enough that a
-    /// terminal `git config user.email …` shows up while the user is still looking at the popover.
+    /// **The fields re-read the config at 2 s while the surface is visible** (06 ▸ Interaction with
+    /// external writers, blessed 2026-07-31): "the watcher never delivers `.git`, so no board event
+    /// can carry a terminal-side config edit — the unfocused-resync courtesy needs its own signal, and
+    /// a visibility-scoped poll is the 15 s paused-state re-read's shape at sheet cadence (a focused
+    /// field keeps its keystrokes; dismissing the sheet stops the poll)."
     private static let pollInterval: Duration = .seconds(2)
 
     var body: some View {

Kanban/UI/Board/BoardInfoPopover.swift

@@ -424,6 +424,13 @@ private struct BoardGitAddAction: View {
                     .fixedSize(horizontal: false, vertical: true)
             }
         }
+        // **The form add-git answers at** (06 ▸ Interaction with external writers, ruled 2026-07-31 —
+        // "Form-anchored operations answer at the form first"; its container moved to the board
+        // settings sheet in the 2026-07-31 split, and these two lines are what that sheet re-points).
+        // Appearing claims the inline surface; disappearing gives it up, which both dismisses the
+        // stale error and sends any answer still in flight to the banner instead of to nobody.
+        .onAppear { git.noteFormVisible(true) }
+        .onDisappear { git.noteFormVisible(false) }
     }
 }
 

Kanban/UI/Card/CardBodyEditSession.swift

@@ -43,12 +43,18 @@ import Observation
 /// `NSUndoManager` of its own, which is what makes it session-scoped). What lives here is the other
 /// half of 05 ▸ Edit's undo sentence: the **session**, whose end is the effective Save.
 /// `endEditSession()` is that moment — the Edit→Preview flip, raw-source entry, or the window
-/// closing — and it is deliberately a named call rather than a side effect of `flush()`, because
-/// pro-m1's auto-commit coalesces exactly here: every debounced tick inside one session rides its
-/// own `performWrite` bracket, and the committer's rule is one commit per *session*, "never per save
-/// tick" (06-history-undo.md ▸ Rules ▸ Auto-commit). In the free tier there is no committer, so
-/// the two calls do the same work today; the seam is what keeps them from having to be pulled apart
-/// later.
+/// closing — and it is deliberately a named call rather than a side effect of `flush()`, because it
+/// is where this session's **one undo step** is registered: every debounced tick inside a session
+/// rides its own `performWrite` bracket, and 13-native-undo.md's rule is one step per session, never
+/// per save tick (`registerUndo`).
+///
+/// **The commit boundary is no longer here** (06-history-undo.md ▸ Rules ▸ Auto-commit, widened
+/// 2026-07-31 — "Board history sees card-window sessions, not gestures"): the committer's
+/// stage-around used to open and close with this flip, and it now opens with the *window* and
+/// releases when the window's session ends, so a comment posted with the body in Preview is inside
+/// the same one commit as the body. What this type still owes that model is `isEditing`, read as
+/// unsaved-content state rather than announced (`CardWindowHost`); the announcement seam it used to
+/// wire went with the widening.
 @MainActor
 @Observable
 public final class CardBodyEditSession {
@@ -110,23 +116,6 @@ public final class CardBodyEditSession {
     @ObservationIgnored
     public var registerUndo: ((_ priorBody: String, _ newBody: String) -> Void)?
 
-    /// **The session boundary, announced** — called with `true` when an Edit session opens and
-    /// `false` when it ends, and with nothing in between.
-    ///
-    /// **No longer the committer's stage-around boundary** (06-history-undo.md ▸ Rules ▸ Auto-commit,
-    /// widened 2026-07-31 — "Board history sees card-window sessions, not gestures"): the exclusion
-    /// used to open and close with this flip, and it now opens with the *window* and releases when its
-    /// session ends, so a comment posted with the body in Preview is inside the same one commit as the
-    /// body. `CardWindowHost` therefore no longer wires this to anything, and the Edit→Preview flip is
-    /// a save point rather than a commit point.
-    ///
-    /// The seam stays, unwired, because it is the only announcement of the boundary this type makes
-    /// and the ordering it carries — that the flush precedes the announcement — is a property worth
-    /// keeping proved (`AutoCommitTests`). A closure for `save`'s reason exactly: this type is a
-    /// buffer and a clock, and it stays testable by having no idea what a repository is.
-    @ObservationIgnored
-    public var editSessionDidChange: ((_ isEditing: Bool) -> Void)?
-
     /// What disk said before this session's **first** landed save — the step's before-value, held
     /// from the first write until the session ends.
     ///
@@ -194,16 +183,16 @@ public final class CardBodyEditSession {
     /// because its card's body was empty (`CardBodyMode.opening(body:)`).
     ///
     /// Idempotent, because the mode can be re-asserted by a menu validation pass or a re-published
-    /// focus value, and a second announcement would register a session that is already registered.
+    /// focus value, and `isEditing` is observed: assigning `true` over `true` would still notify, and
+    /// the notification is a view rebuild for a session that never changed (`adopt(diskBody:)`'s
+    /// equal-assignment note, one property along).
     public func beginEditSession() {
         guard !isEditing else { return }
         isEditing = true
-        editSessionDidChange?(true)
     }
 
     /// The end of one Edit session — the flip back to Preview, raw-source entry, or the window
-    /// closing. Flushes, and marks the boundary pro-m1's auto-commit coalesces on (see the type's
-    /// doc comment).
+    /// closing. Flushes, then registers the session's one undo step (see the type's doc comment).
     @discardableResult
     public func endEditSession() -> CardBodyWriteOutcome {
         let outcome = flush()
@@ -216,14 +205,9 @@ public final class CardBodyEditSession {
             registerUndo?(origin, disk)
         }
         sessionOriginBody = nil
-        // **Last**, after the flush and after the undo step: the committer's nudge must find the
-        // session's final bytes already on disk, or the commit it arms would carry the file as it
-        // stood one keystroke ago. Guarded on `isEditing` so a window closing from Preview — which
-        // calls this too, and should — announces nothing.
-        if isEditing {
-            isEditing = false
-            editSessionDidChange?(false)
-        }
+        // Guarded on `isEditing` for `beginEditSession`'s reason — a window closing from Preview
+        // calls this too, and should, but has no session flag to move.
+        if isEditing { isEditing = false }
         return outcome
     }
 
@@ -243,10 +227,7 @@ public final class CardBodyEditSession {
         cancelPending()
         text = disk
         sessionOriginBody = nil
-        if isEditing {
-            isEditing = false
-            editSessionDidChange?(false)
-        }
+        if isEditing { isEditing = false }
     }
 
     /// `DirtyBufferGuard`'s `attemptSave`: the same flush, with a real failure raised instead of

Kanban/UI/Card/CardComments.swift

@@ -105,6 +105,14 @@ public final class CardComments {
 
     /// The crash-residue sweep, run once when the window opens —
     /// `BoardStore.sweepCommentTrashResidue(inCard:)`.
+    ///
+    /// **It is a sweep of *residue*, not of the trash** (13-native-undo.md ▸ Interaction with the
+    /// trash, ruled 2026-07-31): content a live board step still restores from is that step's backing
+    /// and is left standing, so reopening a card can never destroy the last session's undo. **Nor the
+    /// bytes a collision left**, since the same day's skip-purge decoupling: a coarse step popped as
+    /// stale applied nothing and destroys nothing, and what it was holding stands until the board
+    /// session ends. The gate lives store-side, where the stack is; this seam is unchanged and still
+    /// called exactly once, by `open()`.
     @ObservationIgnored
     public var sweepTrashResidue: (() -> Void)?