fix: add max_length support to Gzip/Brotli decoders for urllib3 2.6+

[CalNightingale]

Summary

Fixes #491. Supersedes #492 (stale).

urllib3 >= 2.6.0 now passes max_length to decoder.decompress() and accesses decoder.has_unconsumed_tail. Our custom _GzipDecoder and _BrotliDecoder wrappers didn't accept these, causing TypeError at runtime.

This PR:

• Updates _GzipDecoder.decompress and _BrotliDecoder.decompress to accept and forward max_length, with a TypeError fallback for older urllib3
• Adds has_unconsumed_tail property to _BrotliDecoder (the proxy class that needs it explicitly)
• Applies the same fix to the async _GzipDecoder in google/_async_resumable_media/
• Adds test coverage for the new max_length forwarding, the fallback path, and has_unconsumed_tail

Feedback from #492 addressed

• max_length=-1 default (chandra-siri) — uses -1 to match urllib3's default, not None
• has_unconsumed_tail only on _BrotliDecoder (chandra-siri) — _GzipDecoder inherits this from the parent urllib3.response.GzipDecoder, so no override is needed
• Lint failure (BrennaEpp) — no trailing whitespace
• Missing test coverage for fallback paths (BrennaEpp) — added tests for max_length forwarding, TypeError fallback, and has_unconsumed_tail (with AttributeError fallback for older urllib3)

Test plan

• pytest tests/unit/requests/test_download.py — all pass
• pytest tests_async/unit/requests/test_download.py — all pass
• Full unit suite (514 tests) — all pass
• No lint issues (flake8 clean)

🤖 Generated with Claude Code

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.