docs(blog): update CI performance gate story

[TimeToBuildBob]

What

Update the existing CI performance-gate article with the September recurrence in the same gptme conversation-list test suite. The cold-scan threshold had drifted 500→600→650ms and still held master red on a healthy path.

The update documents the replacement from gptme/gptme#3900:

• N=25 versus N=100 scaling check
• loose 5s catastrophe bound
• explicit distinction between algorithmic behavior and a hang detector

This extends the existing artifact instead of publishing a duplicate post about the same principle.

Verification

• scoped content validation and URL checks passed
• OG image verified up to date
• full Jekyll build passed
• website pre-commit hooks passed
• PR quality gate: 100/100

[TimeToBuildBob]

🤖 AI code review

Safe to merge — no P0/P1 findings

Confidence 5/5

✅ No findings. The diff looks correct to me on this pass.

Previous review passes

commit	score	findings	engine	when
c26d35d89359	5/5	0	agent	2026-09-21 06:47 UTC

_{Reviewed c26d35d89359 · openai-subscription/gpt-5.6-sol · agent engine · 90s · about this reviewer}

_{Maintainer commands}

_{@TimeToBuildBob review (own line) — fresh review · @TimeToBuildBob fix — a worker acts on the findings. Once per comment; 👀 = received.}