Forbid include-shard file-size bypasses in PMM policy

[konard]

Summary

• Add canonical transformation-rule text forbidding include/pmm/** header surface reduction through .inc, .inl, .ipp, or similar textual include shards.
• Clarify that file-size pressure should lead to compaction or real extraction into a normal .h module with its own responsibility.
• Add repo-policy.json forbidden path globs for new include/pmm/**/*.inc, include/pmm/**/*.inl, and include/pmm/**/*.ipp files.

Verification

• bash scripts/check-docs-consistency.sh
• bash scripts/check-repo-guard-rollout.sh
• git diff --check
• python3 -m json.tool repo-policy.json >/dev/null

Fixes #293

[konard]

Solution summary

Implemented and pushed the governance-only fix for issue 293.

PR: #294

Changed:

• docs/pmm_transformation_rules.md: added canonical rule that .inc / .inl / .ipp include-sharding under include/pmm/** is not compaction and is a policy violation.
• CONTRIBUTING.md: clarified that the 1500-line file-size limit must drive compaction or real .h module extraction, not include shards.
• repo-policy.json: added forbidden path globs for include/pmm/**/*.inc, include/pmm/**/*.inl, and include/pmm/**/*.ipp.

Verification run locally:

• bash scripts/check-docs-consistency.sh
• bash scripts/check-repo-guard-rollout.sh
• git diff --check
• python3 -m json.tool repo-policy.json >/dev/null

Committed as 4f0ee47 Forbid PMM include shard policy bypass, pushed to issue-293-8159ba0e9788, updated PR 294 title/body, and marked it ready for review. Remote checks had started; docs consistency and several quick checks were already passing, while the full CI matrix was still in progress at the time of my last check.

Useful next steps:

1. Watch PR 294 until the remaining CI jobs complete.
2. If repo-guard reports advisory findings, inspect whether it is flagging the intended new forbidden path policy or unrelated existing branch placeholder history.

---

This summary was automatically extracted from the AI working session output.

[konard]

🤖 Solution Draft Log

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Model: GPT-5.4
• Provider: OpenAI
• Public pricing estimate: $0.619193

📊 Context and tokens usage:

• 666.6K / 1.1M (63%) input tokens, 4.7K / 128K (4%) output tokens

Total: (39.8K + 626.8K cached) input tokens, 4.7K output tokens, $0.619193 cost

🤖 Models used:

• Tool: OpenAI Codex
• Requested: gpt-5.4
• Model: GPT-5.4 (gpt-5.4)

📎 Log file uploaded as Gist (7223KB)

• View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.

[konard]

✅ Ready to merge

This pull request is now ready to be merged:

• All CI checks have passed
• No merge conflicts
• No pending changes

---

Monitored by hive-mind with --auto-restart-until-mergeable flag