[engine-failure] PROJ-818-llmxive-follow-up-extending-co-evolving @ tasks: RuntimeError

Automated report (spec 023 / FR-016) — an unexpected engine failure occurred while processing **PROJ-818-llmxive-follow-up-extending-co-evolving** at stage `tasks`.

```
RuntimeError: LLMReviewer[constraint_preservation]: review frontmatter has neither `verdict:` nor `concerns:`/`action_items:` — a contentless/truncated review MUST NOT count as acceptance (false-convergence guard). First 200 chars: '```yaml\n---\nreviewer_name: constraint_preservation\nreviewer_kind: llm\nstage: tasked\nartifact_path: projects/PROJ-818-llmxive-follow-up-extending-co-evolving/specs/001-llmxive-follow-up-extending-co-ev'
```

The project remains schedulable; it will retry on its normal cadence and recover once the underlying defect is fixed. Closing this issue does not require any project-state change.

Start by tracing the tasks-stage handling for LLMReviewer[constraint_preservation] and inspect the review artifact under projects/PROJ-818-llmxive-follow-up-extending-co-evolving/specs/001-llmxive-follow-up-extending-co-evolving. Reproduce the reported RuntimeError and determine why the review frontmatter lacks verdict or concerns/action_items; done means the contentless review is rejected without preventing normal retry recovery.

Written by the indexing model from the issue text.