[engine-failure] PROJ-901-llmxive-follow-up-extending-adaplanbench @ tasks: RuntimeError

Automated report (spec 023 / FR-016) — an unexpected engine failure occurred while processing **PROJ-901-llmxive-follow-up-extending-adaplanbench** 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: tasks\nartifact_path: projects/PROJ-901-llmxive-follow-up-extending-adaplanbench/specs/001-llmxive-follow-up-extending-adapl'
```

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.

Search for the LLMReviewer constraint_preservation handling and the reported false-convergence error, starting with the review frontmatter shown in the report. Trace how a truncated review reaches the tasks stage and verify that contentless reviews are rejected while reviews containing verdict or concerns/action_items are handled correctly.

Written by the indexing model from the issue text.