docs(spec): name the recording comment and fix three prose claims (review F1-F4)

Four prose corrections from the delta contract review's non-blocking findings.
No behaviour change, no contract change, no new argument.

F1 — ADR-0058 Amendment II.3's ruling paragraph cited the date and the decision
batch but not the recording comment. It now names comment `5560086928`, the
comment on this card that records the maintainer reply the block quotes.

F2 — round 3's rewrap left a stub line (`matched). So an`) mid-paragraph. The
paragraph is rewrapped to the block's own idiom; the prose is word-identical.

F3 — the block said it amends D3's closing SENTENCE. It amends the bullet's
last two: the "rewrite *conditioned* on the row is out of contract" sentence is
superseded for the in-place / same-key-set case alongside the "not so a rewrite
can be aimed" one. Now "closing sentences".

F4 — the changeset attributed the ruling to the director seat. The MAINTAINER
ruled; the director seat recorded it. This text ships to consumers inside the
package's CHANGELOG.md, so the misattribution was published. Now "Maintainer
ruling (recorded by the director seat, decision batch #59, 2026-09-06)".

Level re-derived rather than inherited: `packages/spec`'s files[] carries
src/**/*.zod.ts and dist, while docs/adr/** is in no package's files[]. This
round moves no published carrier and changes nothing behavioural, so the
existing `@objectstack/spec: minor` stands.

Claude-Session: https://claude.ai/code/session_01MkQhmuuJAVDjmeWNixwDDH
Co-authored-by: Claude <noreply@anthropic.com>

Author: claude

.changeset/hook-previous-row-invariant-rewrite.md

@@ -4,7 +4,7 @@
 
 feat(spec): `HookContext` admits a row-invariant-in-effect rewrite by per-row `previous` on a predicate write, kept safe by the engine's key-divergence refusal (#16074)
 
-The `hook.zod.ts` contract said that on a predicate (`multi: true`) write the per-row `previous` is supplied *so a guard can REFUSE (throw), not so a rewrite can be aimed*. Three shipped `beforeUpdate` provenance stamps (`sys_email_template`, `sys_sharing_rule`, `sys_webhook`) read `ctx.previous` per row and write `customized: true` conditioned on it — inside the letter of what the engine allows, outside the stated purpose of the input they use. Director ruling, decision batch #59 (2026-09-06), option 1: **the contract admits the shape.**
+The `hook.zod.ts` contract said that on a predicate (`multi: true`) write the per-row `previous` is supplied *so a guard can REFUSE (throw), not so a rewrite can be aimed*. Three shipped `beforeUpdate` provenance stamps (`sys_email_template`, `sys_sharing_rule`, `sys_webhook`) read `ctx.previous` per row and write `customized: true` conditioned on it — inside the letter of what the engine allows, outside the stated purpose of the input they use. Maintainer ruling (recorded by the director seat, decision batch #59, 2026-09-06), option 1: **the contract admits the shape.**
 
 The amended D3 clause (`HookContextSchema.input` TSDoc, mirrored in `bulk-write-hook-conformance.ts`) now states:
 

docs/adr/0058-expression-and-predicate-surface.md

@@ -442,15 +442,16 @@
 > **Amendment II.3 (2026-09, #16074 maintainer ruling, decision batch #59) —
 > a ROW-INVARIANT-IN-EFFECT rewrite is ADMITTED, and the ENGINE is what makes
 > it safe.**
-> _Amends D3's closing sentence in Addendum II above by ADDITION: the 2026-08
+> _Amends D3's closing sentences in Addendum II above by ADDITION: the 2026-08
 > text is left standing as the dated record of what was decided then, carrying
 > a forward pointer to here. This block RECORDS a decision already taken; it
 > takes none._
 >
 > **The ruling.** Maintainer reply, verbatim: 「16063 c,  其他同意」
-> (2026-09-06, decision batch #59, on this card #16074). Option 1 adopted: the
-> CONTRACT admits the shape. The three shipped provenance stamps are made
-> conforming by amending the contract, not by being changed to fit it.
+> (2026-09-06, decision batch #59, recorded on this card #16074 in comment
+> `5560086928`). Option 1 adopted: the CONTRACT admits the shape. The three
+> shipped provenance stamps are made conforming by amending the contract, not
+> by being changed to fit it.
 >
 > **What the rule now is.** On a predicate write (`multi: true`), per-row
 > `previous` is supplied so a guard can REFUSE (throw) **and** so a `before*`
@@ -473,9 +474,9 @@
 > batch is not judged at all. To an operator that refusal is an ADR-0112
 > envelope, `status: 400`, carrying `keys` (the sorted keys some rows' hooks
 > wrote and other rows' did not) and `rows` (how many rows the predicate
-> matched). So an
-> author does not have to be TRUSTED to be row-invariant; a hook that is not
-> gets a loud, whole-batch 400 instead of a half-stamped table.
+> matched). So an author does not have to be TRUSTED to be row-invariant; a
+> hook that is not gets a loud, whole-batch 400 instead of a half-stamped
+> table.
 >
 > **The three shapes the rule does NOT admit.**
 >