fix(ci): extend a2ui production smoke timeout

[blove]

Summary

• extend the production send/receive timeout budget for the chat/a2ui smoke
• keep the existing 30s response budget for the simpler streaming and planning smoke checks
• set the Playwright test timeout above the per-response wait so the assertion can use its full budget

Root Cause

After #397 fixed the shared backend verifier timeout, the next production smoke step reached Playwright and failed only on chat/a2ui. That smoke uses the GPT-5 structured-output A2UI graph and timed out at Playwright's default 30s test limit while waiting for a2ui-surface. The assertion also had a 30s timeout, so the test-level timer could kill it before the A2UI surface had enough time to render.

Validation

• BASE_URL=https://cockpit.cacheplane.ai EXAMPLES_URL=https://examples.cacheplane.ai node_modules/.bin/playwright test apps/cockpit/e2e/production-smoke.spec.ts --reporter=list locally: 32 passed, 3 skipped because local shell has no OPENAI_API_KEY
• git diff --check

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
cacheplane	Ready	Preview, Comment	May 17, 2026 5:06am