fix(pm): drop the stale root-walk residue row for check:pm-half-states

CI red at `Lint & Repo Gates` step 32 (run 34759389219): `dispatch-gates
self-test: 1 of 1682 case(s) failed` — 「every gate CI runs whose own source
sweeps the repo root is DECLARED whole-tree, declared path-less, derivable by
path, or a justified row in ROOT_WALK_RESIDUE_LEDGER — listed but no longer a
member: check:pm-half-states」.

Root cause, measured: this PR's self-test reads two sibling sources as program
text — `sweep-stale-finding.mjs`, to pin that it still ALIASES the one
stale-`finding` screen rather than re-growing a copy, and `check-prior-rulings.
mjs`, to pin that the prior-ruling line's writer still prints the key the patrol
greps. The derivation reads those literals and now places the family BY PATH
(`dispatch-gates scripts/pm/sweep-stale-finding.mjs` ⇢ 「program text read by
scripts/pm/check-half-states.mjs」), so it left the residue population — and the
ledger reds on a listed family that is no longer a member, by its own contract:
a stale exclusion is an exclusion nobody measures.

The repair is the ledger's, ⛔ not the self-test's: the row goes, with a note
saying why the family left and why this placement is the OPPOSITE of the #15753
one the row was written about — that one came from a noise-floor constant and
said the opposite of what the constant declares, while these two literals are
exactly what the self-test reads. A card touching either sibling now really does
owe this gate.

`dispatch-gates --self-test`: 1682 of 1682 pass, exit 0.

Claude-Session: https://claude.ai/code/session_01DAcomhvR9kKizeYgg89Vo8
Co-authored-by: Claude <noreply@anthropic.com>

Author: claude

scripts/pm/dispatch-gates.mjs

@@ -3227,15 +3227,18 @@ export const ROOT_WALK_RESIDUE_LEDGER = [
       + '(the pnpm script would drag the scan in with it). The repo-root default parameter belongs to the scan half '
       + 'CI does not schedule, so a whole-tree row for this family would advertise work no workflow performs.',
   ],
-  [
-    'check:pm-half-states',
-    'lint.yml runs the SELF-TEST HALF and never the sweep — the npm script is `check-half-states.mjs '
-      + '--self-test`. The `git ls-files` oracle that vouches for the root walk is read ONCE PER LIVE SWEEP, after '
-      + 'the board gathering, so it belongs to the networked half-state patrol that no lint job schedules; a '
-      + 'whole-tree row here would advertise work this family never performs. It became a member under #15753: it '
-      + 'used to be placed BY PATH on every changeset card, through the gate\'s own noise-floor constant read as a '
-      + 'watch surface — a placement that said the opposite of what the constant declares.',
-  ],
+  // ⚖️ `check:pm-half-states` LEFT this population under #16904's D2 and its row
+  // is gone with it — a listed family that stops being a member reds here, and
+  // a stale exclusion is an exclusion nobody measures. It is now placeable BY
+  // PATH, and truthfully: `check-half-states.mjs`'s self-test reads two sibling
+  // sources as program text — `scripts/pm/sweep-stale-finding.mjs` and
+  // `scripts/pm/check-prior-rulings.mjs` — to pin that the sweep still ALIASES
+  // the one stale-`finding` screen rather than re-growing a copy, and that the
+  // prior-ruling line's writer still prints the key the patrol greps. So a card
+  // touching either sibling really does owe this gate, which is the opposite of
+  // the #15753 placement this row was written about: that one came from a
+  // noise-floor constant and said the opposite of what the constant declares,
+  // while these two literals are exactly what the self-test reads.
   [
     'scripts/symbol-anchors.mjs --self-test',
     'a shared grammar-and-extractor LIBRARY, invoked by CI only as its own self-test; its `git ls-files` runs over a '