[engine-failure] PROJ-362-evaluating-the-statistical-validity-of-c @ tasks: RuntimeError

Automated report (spec 023 / FR-016) — an unexpected engine failure occurred while processing **PROJ-362-evaluating-the-statistical-validity-of-c** at stage `tasks`.

```
RuntimeError: LLMReviewer[ordering]: 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: ordering\nreviewer_kind: llm\nstage: tasked\nartifact_path: projects/PROJ-362-evaluating-the-statistical-validity-of-c/specs/001-evaluating-the-statistical-validity-of-c/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 LLMReviewer[ordering] validation at the tasks stage and inspect the truncated review artifact under projects/PROJ-362-evaluating-the-statistical-validity-of-c/specs/001-evaluating-the-statistical-validity-of-c/tasks. Determine why the review frontmatter lacks verdict or concerns/action_items, then verify that contentless reviews cannot count as acceptance and the project retry can recover.

Written by the indexing model from the issue text.