Cover proposal 0021 transition observation

[SoundBlaster]

Summary

• What changed? Updated Proposal 0021 observation coverage to track the current supervisor JSON emission path.
• Link related issue/task (if any): graph next-move reobservation follow-up.

Motivation

• Why is this change needed? Proposal 0021 still pointed at the old direct JSON print marker, so derived proposal coverage reported a stale reobservation gap.
• What problem, gap, or user need does it address? Keeps the proposal runtime registry aligned with the implemented supervisor output-routing behavior.

Goals

• What should this PR achieve? Mark Proposal 0021 observation coverage as covered and guard it with a focused regression test.
• What is intentionally out of scope? No runtime behavior changes and no canonical spec changes.

Changes

• Replaced the stale Proposal 0021 observation marker with the current emit_supervisor_json(... normalized_output_mode) marker.
• Added a regression asserting Proposal 0021 has implemented runtime, covered validation, covered observation, and no next gap.

Validation

• Tests added/updated for changed behavior
• Local checks passed

Commands run:

python3 -m pytest -q tests/test_supervisor.py -k 'proposal_0021 or proposal_runtime_index'
git diff --check
make backlog
make next-move
python3 -m pytest -q tests/test_supervisor.py

Results:

• Focused proposal runtime tests: 3 passed, 605 deselected.
• Full supervisor tests: 608 passed in 285.25s.
• Backlog now advances from Proposal 0021 to the next reobservation gap.

Risks / Notes

• Backward compatibility impact: none; this changes registry observation metadata only.
• Migration/config changes required: none.
• Known limitations: Proposal 0036 remains as the next reobservation gap and is handled in the follow-up stacked PR.

Checklist

• PR title clearly describes the change
• Scope is focused and minimal
• Documentation updated (or N/A)
• No secrets or sensitive data added

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: fc9f0d5780

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[SoundBlaster]

Blocking review finding: the new observation marker for Proposal 0021 is too broad and can produce false coverage.

tools/proposal_runtime_registry.json now uses emit_supervisor_json(report, output_mode=normalized_output_mode) as an observation marker. The marker evaluator is a plain substring check over the whole file, and this call shape appears in many standalone report-producing paths, not only in the --validate-transition-packet path. That means observation_coverage can remain covered even if the transition-packet command stops routing the validator report correctly.

Please make this marker transition-specific. For example, use a marker around report = validate_transition_packet_file( or the standalone-command guard text for --validate-transition-packet, so the registry observes the actual Proposal 0021 behavior rather than generic JSON output plumbing.

Note: I tried to submit this as a formal request-changes review, but GitHub rejects request-changes reviews on one's own PR, so I am posting it as a blocking PR comment instead.