ci(gate): give each gate part a 45 min exec deadline

[debuggingfuture]

Each gate part gets a 45 min wall clock, up from 30, which matches the dispatcher's 2700 s part deadline (fractalboxdev/flare-dispatch#198).

Problem & Insight

• features.binary-*, workspace.cli, evaluate and budget.full take 1250–1750 s on a passing run. Under the 1800 s deadline most of them end in ExecTimeout with no verdict, so the parent check reads red on every pull request.
• Splitting those parts makes the gate slower overall. The dispatcher runs parts two at a time, and each added part repeats a 550–760 s cold build.
• The ADR row "Workflow steps admit at most 30 minutes" doesn't hold: the check run's exec is detached and polled across steps, so the deadline is set by the dispatcher, not by the platform.

Take

• assurance.gate.container / assurance-container-wall-clock: 45 min.
• A-assurance records the decision and its options: a longer deadline, splitting by test module, or a shared build cache. It also corrects the stale platform-limit row.
• AGENTS.md: a check still running 60 min after dispatch counts as stalled.

Todos

• fix(contextful-gate): give each part a 2700 s exec deadline flare-dispatch#198 deployed
• Three consecutive pull requests with no ExecTimeout

Closes #465

[cloudflare-workers-and-pages]

🚀 Deploying Preview to Cloudflare 🚀

Preview URL: https://ci-part-cap-2700.spec.contextful.work (commit ded6436)

This URL reflects your latest Preview deployment

Preview Deployments by commit

Status	Deployment URL	Commit	Updated (UTC)	See this deployment's details
Build: Success ✅ Deployment: Success ✅ View logs ↗	https://6c013dca.spec.contextful.work	ded6436	2026-10-11T16:03:00.116Z	Visit the dashboard ↗
Build: Success ✅ Deployment: Success ✅ View logs ↗	https://1cd404d9.spec.contextful.work	c33c33c	2026-10-11T15:39:36.457Z	Visit the dashboard ↗
Build: Success ✅ Deployment: Success ✅ View logs ↗	https://f9df7ef3.spec.contextful.work	7c95652	2026-10-11T15:33:29.802Z	Visit the dashboard ↗

[flaredispatch-fractalboxdev]

⚠️ pr-review could not complete: { "step": "review", "cause": Error: ModelCallFailed: openai returned 429: {"success":false,"result":[],"messages":[],"error":[{"code":2003,"message":"Rate limited"}],"name":"AiGatewayError","httpCode":429,"internalCode":2003,"message":"Rate limited","description":"Rate limited"}, "summaryMd": undefined }

📋 View full logs & reviewed diff ↗