[engine-failure] PROJ-482-assessing-the-sensitivity-of-common-stat @ tasks: RuntimeError

Automated report (spec 023 / FR-016) — an unexpected engine failure occurred while processing **PROJ-482-assessing-the-sensitivity-of-common-stat** 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-482-assessing-the-sensitivity-of-common-stat/specs/001-assessing-the-sensitivity-of-common-stat/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 at the tasks-stage LLMReviewer[coverage] handling for spec 023 / FR-016 and inspect how review frontmatter is parsed. Reproduce the retry around the reported RuntimeError; done means a truncated or contentless review cannot count as acceptance and the project remains schedulable for normal recovery.

Written by the indexing model from the issue text.