Copy chunks as MaybeUninit<u8>

[folkertdev]

When the output buffer contained uninitialized memory, MIRI reported an error in the copy_chunk_unchecked function. That is unsound, though I don't see any way that could have been observed or exploited.

The bounds checks are all valid as far as I can see, so I think it's OK to just read/write the memory without assuming that it is initialized. Let's see if this has a performance impact though.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Flag	Coverage Δ
fuzz-compress	41.78% <100.00%> (ø)
fuzz-decompress	30.15% <100.00%> (-0.03%)	⬇️
test-aarch64-apple-darwin	89.47% <100.00%> (+0.01%)	⬆️
test-x86_64-apple-darwin	87.70% <100.00%> (+0.03%)	⬆️
test-x86_64-unknown-linux-gnu	90.16% <100.00%> (+0.09%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
zlib-rs/src/inflate/writer.rs	91.54% <100.00%> (ø)

... and 4 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.

[folkertdev]

perf looks ok to me https://github.com/trifectatechfoundation/zlib-rs/actions/runs/14524839981