fix: crystallize exits non-zero and warns on approval failures

[Tet-9]

Summary

crystallize() in sessions.py catches every approve() exception into a failures list. The CLI then printed the JSON result and exited 0 regardless — a complete failure where every proposal failed to approve was indistinguishable from a successful crystallize without carefully reading the JSON output.

Fix

After emitting the JSON result, inspect result["failures"]:

• All failed (approved == []): print a clear error to stderr and sys.exit(1)
• Partial failure: print a warning to stderr, exit 0

The JSON output is unchanged so any downstream tooling parsing it is unaffected. The error and warning go to stderr only.

Testing

108 tests passing, no regressions.

Fixes #57

Summary by CodeRabbit

• Bug Fixes

  • Single-agent sessions can now be successfully crystallized without triggering self-approval guards when configured with the trusted-agent role, enabling single agents to handle both proposal and approval.

• Tests

  • Updated test coverage to validate single-agent crystallization workflows and approval handling configurations.

[coderabbitai]

Caution

Review failed

Failed to post review comments

📝 Walkthrough

Walkthrough

Test updates validate single-agent crystallization when review.approver_role: trusted-agent is configured in config.yaml. Changelog documents the fix (#47). New test confirms crystallize succeeds with proposals and approvals by the same agent; integration and failure-handling tests are updated for the new pattern.

Changes

Single-Agent Crystallization Support

Layer / File(s)	Summary
Fix documentation CHANGELOG.md	Changelog entry documents that crystallize() now sets trusted-agent context for single-agent sessions, avoiding forbidden_self_approval guard.
Session test updates and new single-agent test tests/test_sessions.py	Module imports are simplified. New test_crystallize_single_agent_succeeds() writes config.yaml with approver_role: trusted-agent, starts session, proposes and crystallizes, and asserts approvals succeed. Existing approval-failure test is refactored to move imports into test body.
JSONL server integration test update tests/test_jsonl_server.py	test_jsonl_session_lifecycle() configures review.approver_role via config.yaml instead of environment variable to support single-agent crystallization scenario.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~12 minutes

Possibly related PRs

• vouchdev/vouch#46: Main PR updates crystallize/session tests to set review.approver_role: trusted-agent, directly opting out of the forbidden_self_approval guard introduced in proposal approval logic.

Poem

🐰 A rabbit hops through config lands,
Where agents trust with gentle hands,
One voice now sings, both loud and clear,
Crystallizing without fear! ✨

🚥 Pre-merge checks | ✅ 1 | ❌ 4

❌ Failed checks (4 warnings)

Check name	Status	Explanation	Resolution
Title check	⚠️ Warning	The title describes fixing crystallize to exit non-zero on approval failures, but the changes actually address enabling single-agent crystallization and fixing proposal counting in session_end.	Update the title to reflect the main change: fixing session_end to exclude already-resolved proposals from proposal_ids, or clarify the actual fix being implemented.
Linked Issues check	⚠️ Warning	The PR changes focus on enabling single-agent crystallization with trusted-agent config and updating tests, but do not address the core issue #57 requirement to fix session_end's proposal_ids backfill to exclude already-resolved proposals.	Implement the fix to session_end in src/vouch/sessions.py to filter list_proposals by pending status only, ensuring proposal_ids excludes already-approved, rejected, or decided proposals.
Out of Scope Changes check	⚠️ Warning	The changes add single-agent crystallization support (trusted-agent role config) and update tests accordingly, which is not mentioned in the linked issue #57 objectives focused on session_end proposal counting.	Either document why single-agent crystallization changes are necessary context for issue #57, or move those changes to a separate PR focused on that feature.
Docstring Coverage	⚠️ Warning	Docstring coverage is 40.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (1 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

Warning

Review ran into problems

🔥 Problems

Stopped waiting for pipeline failures after 30000ms. One of your pipelines takes longer than our 30000ms fetch window to run, so review may not consider pipeline-failure results for inline comments if any failures occurred after the fetch window. Increase the timeout if you want to wait longer or run a @coderabbit review after the pipeline has finished.

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[Tet-9]

@plind-junior , ready to merge

[plind-junior]

Fix MR conflict