[engine-failure] PROJ-506-predicting-material-stiffness-from-micro @ tasks: RuntimeError

Automated report (spec 023 / FR-016) — an unexpected engine failure occurred while processing **PROJ-506-predicting-material-stiffness-from-micro** at stage `tasks`.

```
RuntimeError: LLMReviewer[executability]: 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: executability\nreviewer_kind: llm\nstage: tasked\nartifact_path: projects/PROJ-506-predicting-material-stiffness-from-micro/specs/001-predicting-material-stiffness-from-micro/t'
```

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[executability] validation at the `tasks` stage and inspect the review frontmatter shown in the report for PROJ-506. Reproduce the retry if possible; done means a contentless or truncated review cannot count as acceptance, while a valid review with `verdict:` or `concerns:`/`action_items:` can proceed without the RuntimeError.

Written by the indexing model from the issue text.