test: make http2 close-while-writing single-shot

[trivikr]

This makes test-http2-close-while-writing handle only the first server-side
request data chunk.

The test only needs one chunk to exercise the regression path: the server queues
a write, then the client stream is destroyed on the next tick. Keeping the data
listener active can queue multiple writes and schedule multiple destroys for the
same request body, which makes the test occasionally hang until the harness
timeout.

Refs: https://github.com/nodejs/reliability/blob/main/reports/2026-05-17.md#jstest-failure

---

Assisted-by: openai:gpt-5.5

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.05%. Comparing base (a9b0a83) to head (06dd2a9).
⚠️ Report is 6 commits behind head on main.

Additional details and impacted files

@@ Coverage Diff @@## main #63388 +/- ##
==========================================
- Coverage 90.06% 90.05% -0.01% 
==========================================
Files 714 714 Lines 225629 225648 +19 Branches 42702 42697 -5 ==========================================
+ Hits 203211 203215 +4 + Misses 14232 14219 -13 - Partials 8186 8214 +28 

see 59 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[aduh95]

Keeping the data
listener active can queue multiple writes and schedule multiple destroys for the
same request body, which makes the test occasionally hang until the harness
timeout.

I don't really follow the logic, how scheduling multiple destroy would result in a hang? That sounds like a bug

[lpinca]

#58252 (comment)

[trivikr]

You're right. I should have searched for test-http2-close-while-writing before attempting this PR which would have helped me discover prior attempts. I've been following that in my latest 5 PRs.

[trivikr]

I've posted a bugfix attempt in https://www.github.com/nodejs/node/pull/63439, and will request review after GitHub CI is successful.

[lpinca]

See also #63249

As a happy side bonus, as part of this work I hunted down the underlying issue for #58252: buggy cleanup of reset streams.