[engine-failure] PROJ-251-investigating-the-correlation-between-gu @ tasks: RuntimeError

Automated report (spec 023 / FR-016) — an unexpected engine failure occurred while processing **PROJ-251-investigating-the-correlation-between-gu** at stage `tasks`.

```
RuntimeError: LLMReviewer[coverage]: 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: coverage\nreviewer_kind: llm\nstage: tasked\nartifact_path: projects/PROJ-251-investigating-the-correlation-between-gu/specs/001-investigating-the-correlation-between-gu/tasks.'
```

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 RuntimeError at LLMReviewer[coverage] during the `tasks` stage, using the reported artifact path `projects/PROJ-251-investigating-the-correlation-between-gu/specs/001-investigating-the-correlation-between-gu/tasks.` as the failing entry point. Inspect how review frontmatter is produced and validated; done means a contentless or truncated review cannot count as acceptance and the project remains retryable.

Written by the indexing model from the issue text.