docs: cover reused-process runtimes (AWS Lambda) in deploy.md

[pandayv]

Fixes #3590.

docs/troubleshooting.md documents the single-use StreamableHTTPSessionManager.run() error for two causes: a Mount swallowing a sub-app's lifespan, and several long-running workers without sticky sessions. It doesn't cover a third: a serverless runtime that reuses one warm process across separate invocations (AWS Lambda, in our case). The app built once at import time works on the first invocation and fails every request on the second, once the container is reused, which is the default in production.

This adds that as its own section in deploy.md, right after "Change notifications across replicas," with the per-invocation fix and a short note on why it's doubly true under SnapStart. Also links it from the existing bullet list in troubleshooting.md and adds a matching line to deploy.md's recap.

Built the docs locally with scripts/docs/build.sh, English and all 12 translated sites, 0 warnings.

[github-actions]

This PR has been closed automatically. This repo only keeps pull requests open when they come from a maintainer, or from a contributor a maintainer has assigned to the linked issue, and you aren't currently assigned to #3590.

If a maintainer assigns you to #3590, this PR reopens on its own and there's nothing more you need to do here. Assignment is a maintainer call based on capacity; comments that only ask to be assigned don't factor in. What does help is engaging on the issue itself by confirming the repro, explaining why it matters for your use case, or describing the approach you'd take.

You're welcome to keep pushing commits here (just avoid force-pushing, since GitHub can't reopen a rewritten branch), but that on its own won't get the PR reviewed or the issue assigned, and realistically most auto-closed PRs stay closed. There's no need to open a new PR either way.

CONTRIBUTING.md has the full reasoning, but in short:

• We're a small team with very little capacity to review community PRs right now.
• Many recent PRs are AI-generated with little human review, and reviewing one carefully still costs a maintainer as much time as it ever did. A well-described issue is usually more useful to us than the code.

Maintainers: reopen, remove missing-issue-link, or add bypass-issue-check to override.