quic: fixup quic stream variable chunk len

[jasnell]

Fixes: #63216

[nodejs-github-bot]

Review requested:

• @nodejs/quic

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.03%. Comparing base (facd71e) to head (eafeb01).
⚠️ Report is 39 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #63230      +/-   ##
==========================================
- Coverage   90.04%   90.03%   -0.01%     
==========================================
  Files         713      714       +1     
  Lines      224950   225243     +293     
  Branches    42530    42572      +42     
==========================================
+ Hits       202548   202802     +254     
- Misses      14188    14221      +33     
- Partials     8214     8220       +6     

Files with missing lines	Coverage Δ
lib/internal/quic/quic.js	100.00% <100.00%> (ø)

... and 44 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.

[nodejs-github-bot]

CI: https://ci.nodejs.org/job/node-test-pull-request/73430/ 💛

[pimterry]

This approach looks much better, very nice 👍

I take it this means we're going towards option 1 from my previous backpressure comment? I.e. stream/iter writables are required to always accept one pending write, even in strict mode, and to deal with any internal byte-based limits & backpressure via internal buffering only. Doesn't need to happen immediately, but once we're settled there it'd be good to write that into the stream/iter writer docs explicitly.

[mcollina]

lgtm

[jasnell]

@pimterry ... I think the backpressure strategy for both stream/iter and this as an application of that api still needs a thorough review, which I plan to get to soon.

[jasnell]

Landed in e9c49ea