feat: add fast forge work-record reconciliation

Assistant: codex
Assistant-Model: gpt-5.6-sol
Assistant-Session: 01a053ff-1d6f-7fe2-ac1c-a6eb40a42a0c

docs/cluster-operating-model.md

@@ -138,14 +138,12 @@ Cutover sequence reference: `CUST-WP-0011-T07` (2026-07-03).
 File-backed workplans remain authoritative (ADR-001). After commits:
 
 ```bash
-make fix-consistency REPO=<slug>
-# or from repo root:
-make fix-consistency-here
+uv run --project ~/repo-manager rmgr sync --path /path/to/repo --push
 ```
 
-The 15-minute all-repo sweep is owned by activity-core on Railiance01. It
-reaches the API through the `actcore-state-hub-bridge` proxy chain. Manual
-invocation from the workstation still works:
+The command verifies `primary/railliance01`, then central derives and applies
+the exact pushed Forgejo commit. `statehub fix-consistency` remains the broad
+deep-audit tool; it is not the interactive registration/update path.
 
 ```bash
 curl -s -X POST http://127.0.0.1:8000/consistency/sweep/remote-all \

docs/offline-write-buffer.md

@@ -109,6 +109,7 @@ remain queueable or rejected per the write allowlist above.
 2. Run statehub outbox status on each host that may have queued writes.
 3. Run statehub outbox replay until no due queued envelopes remain.
 4. Review conflict envelopes manually.
-5. Run `statehub fix-consistency` so file-backed workplan/task state
-   remains canonical after replay.
+5. Run `rmgr sync --path <repo> --push` for repositories with file-backed
+   changes. Use `statehub fix-consistency` afterward only if a deep audit is
+   required.
 6. Record a progress note with non-secret replay counts.