zlib: reject reset after gzip emitted incomplete output

[xia-chao]

gzip writes a header on the first write. Calling reset after that starts a new member while that header is already in the output, so gunzip fails.

Same idea as the zstd incomplete-frame case in #66088, but only for gzip. zlib deflate can still reset after a flush: the dictionary test throws the first output away and reuses the compressor. Raw deflate is unchanged.

reset before any write still works. flush then end, without reset, still works.

I ran the new gzip reset test, the dictionary test, the existing reset tests, test-zlib.js, and the flush tests on a Release build.

[codecov]

Codecov Report

❌ Patch coverage is 88.88889% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 90.29%. Comparing base (513981d) to head (96593b0).
⚠️ Report is 3 commits behind head on main.

Files with missing lines	Patch %	Lines
src/node_zlib.cc	88.88%	0 Missing and 2 partials ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main   #66179   +/-   ##
=======================================
  Coverage   90.28%   90.29%           
=======================================
  Files         790      790           
  Lines      272043   272062   +19     
  Branches    51945    51945           
=======================================
+ Hits       245619   245650   +31     
+ Misses      16931    16916   -15     
- Partials     9493     9496    +3     

Files with missing lines	Coverage Δ
src/node_zlib.cc	80.00% <88.88%> (+0.20%)	⬆️

... and 27 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.

[xia-chao]

Thanks for the review! Narrowed this to gzip only in a follow-up commit — could you take another look?