fix(session): requeue Phase 2 on lock contention and pause before requeue

[OnePerson2020]

Description

• LockAcquisitionError in Phase 2 no longer fails the archive: _run_memory_extraction re-raises it without writing .failed.json or failing the task, and resume_queued_commit returns False, so SessionCommitProcessor requeues the message. Completed steps are already checkpointed in completed_memory_steps, so the rerun continues where it stopped.
• SessionCommitProcessor pauses 2 s before re-enqueueing a commit that has to wait (predecessor pending or lock contention), instead of cycling it continuously (the embedding processor already sleeps retry_after before re-enqueueing).

Human Involvement

• A human participated in the implementation or review loop
• This PR was generated entirely by AI agents without human participation in the loop

Related Issue

Fixes #6082

Type of Change

• Bug fix (non-breaking change that fixes an issue)
• New feature (non-breaking change that adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Refactoring (no functional changes)
• Performance improvement
• Test update

Changes Made

• session.py: except LockAcquisitionError before the generic handler in _run_memory_extraction (telemetry + warning, re-raise); resume_queued_commit maps it to False.
• session_commit_processor.py: await asyncio.sleep(_REQUEUE_DELAY_S) (2 s) before the re-enqueue.
• Tests: new test_phase2_lock_contention_requeues_without_failing_archive; test_process_requeues_deferred_commit_and_resets_root_context now also checks the pause comes before the enqueue.

Testing

• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• I have tested this on the following platforms:
  • Linux
  • macOS
  • Windows

echo '{}' > /tmp/min.conf
OPENVIKING_CONFIG_FILE=/tmp/min.conf uv run --extra test pytest tests/unit/session tests/session tests/storage/test_session_commit_processor_identity.py -q --no-cov --ignore=tests/session/test_session_commit_race.py
uv run --extra dev ruff check openviking/session/session.py openviking/storage/queuefs/session_commit_processor.py tests/storage/test_session_commit_processor_identity.py tests/unit/session/test_session_commit_resume.py
uv run --extra dev ruff format --check (same files)
git diff --check

• Without the code change both tests fail (resume_queued_commit returns True after marking the archive failed; no pause before the enqueue); with it they pass.
• Across tests/unit/session, tests/session and the processor tests, the set of other failures is identical before and after the change in my environment (vectordb-engine / full-config tests; test_session_commit_race.py hangs on main locally, so it is excluded). Ruff and git diff --check clean.
• Running in our deployment since 2026-10-09.

Checklist

• My code follows the project's coding style
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• Any dependent changes have been merged and published

Additional Notes

No API or data-format change; no new dependency.

🤖 Generated with Claude Code