fix: a thread counts as synced only when the store confirms its row (GIT-117)

[nTEG-dev]

What

A thread counts as synced only when the store confirms its row (GIT-117).

• Confirmed writes only: the sync loop uses throwing variants of the helpers.
  • A create counts when directUpsert returns the row.
  • A resolve or touch counts only when the PATCH returns a row; 0 rows is a failure.
  • A thread the store doesn't have is a failure, not a no-op.
  • touchThreadsInSupabase now reports per-thread outcomes.
  • The boolean/null wrappers used by create_thread and resolve_thread are unchanged.
• Partial sync: the close header says PARTIAL (the session saved, some threads didn't) and the message names each failed id with its error. threads.json keeps them marked sync_pending. Session files are retained, as before.
• Carry-forward: session_start keeps sync_pending threads whatever their age (they were dropped as "stale" after 24 h). Their local state wins over the store's copy, so an unwritten resolution survives, until a later close writes them. A full sync clears the flag.

Why

The helpers caught their own failures and returned null / false / void, and the loop never read the result. So a thread the store rejected was counted as synced. session_close reported COMPLETE, pruned threads.json and deleted the session files. A local-only thread was then discarded as stale 24 h later, and a failed resolution was overwritten by the store's still-open copy. The GIT-70 test didn't catch it because it failed only the lookup, the one call whose error propagated.

How to test

Venue (vujndhcumgroztfkkdus, v1.8.0 setup.sql; production not touched). A new flow probe makes the store really reject a thread: its source_session names a session the store has never seen, so FK 23503 fires. After the close, it restarts on the same home.

	origin/main 1.11.0	This branch
Store rejected the thread	yes	yes
Close header	COMPLETE	PARTIAL, names t-fk…
threads.json	no pending marker	sync_pending: true
Next session_start	not carried	Carrying forward 1 thread(s) … t-fk…
Exit	1	0

The probe's own intentional 409 is excluded from the failure gate; every other request still counts.

Unit tests: 1363/1363. thread-sync-result.test.ts gains 6 tests that fail the writes (create rejected; resolve 0 rows; resolve rejected; touch rejected; one failure among three; an all-confirmed control). 5 fail on main; the control passes on both.

Existing tests

None changed. The 13 existing sync tests still pass. Their supabase-client mock lacked directPatch, so the resolve and touch paths had never been exercised (the audit's mock-blindness shape); the mock now provides it.

Checklist

• Fails on main, passes on the branch (unit + venue v1.8.0)
• No schema change

🤖 Generated with Claude Code