fix: keep Code Mode tasks running after tool failures (#138044)

* fix: keep Code Mode tasks running after tool failures

Remove forced read-only reconciliation, recovery_resume, and the one-mutation continuation limit. Failed cells return to the normal agent loop so users can inspect partial effects and complete multi-step work with their configured tools. Preserve cancellation, permissions, explicit terminal outcomes, lifecycle receipts, and automatic whole-run replay protection.

* style: format Code Mode callback with pinned oxfmt

* test: update failed-tool QA references after recovery removal

* test(android): observe connection admission after stop

A startup WebSocket request can reach MockWebServer after Stop retires it,
causing the foreground-reentry assertion to report a reconnect that never
happened. Reuse the existing call-through session shadow to observe actual
connection admissions while preserving the real runtime and offline checks.

The same test bytes have a reproduced before failure, a rejected deliberate
reconnect, and 134 passing service/session sibling tests. Android lint passes;
the updated PR CI will verify this candidate before landing.

* test: repair ClawHub QA documentation reference

qa/scenarios/channels/qa-channel-failed-tool-terminal-finalization.yaml

@@ -18,7 +18,8 @@ scenario:
     - The failed tool call is never replayed or described as successful.
     - The model receives one ordinary continuation with its failed result and qa-channel delivers exactly one visible reply.
   codeRefs:
-    - src/agents/embedded-agent-runner/run/code-mode-outcome.ts
+    - packages/agent-core/src/agent-loop.ts
+    - src/agents/embedded-agent-runner/extensions.ts
     - extensions/qa-lab/src/providers/mock-openai/server.ts
   execution:
     kind: flow

qa/scenarios/plugins/clawhub-release-policy-contracts.yaml

@@ -18,7 +18,7 @@ scenario:
     - A committed source change without a version bump fails the ClawHub release check.
     - The same source change passes after the package version and compatibility metadata are bumped and committed.
   docsRefs:
-    - docs/clawhub/publishing.md
+    - docs/plugins/building-plugins.md
     - docs/plugins/community.md
     - docs/help/testing-updates-plugins.md
   codeRefs:

qa/scenarios/scheduling/cron-failed-tool-terminal-finalization.yaml

@@ -21,7 +21,8 @@ scenario:
     - The failed tool never replays and qa-channel receives exactly one final reply.
   codeRefs:
     - src/cron/isolated-agent/run-executor.ts
-    - src/agents/embedded-agent-runner/run/code-mode-outcome.ts
+    - packages/agent-core/src/agent-loop.ts
+    - src/agents/embedded-agent-runner/extensions.ts
     - extensions/qa-lab/src/providers/mock-openai/server.ts
   execution:
     kind: flow