Add Browser Tests and YTS WPT Tests to Presubmit

[Libzu]

Integrate Cobalt Browser tests and YTS WPT tests into the GitHub CI workflow.

Bug: 473909877

[github-actions]

🤖 Gemini Suggested Commit Message

---

ci: Add browser and YTS WPT tests to presubmit

This change integrates Cobalt Browser tests, YTS Web Platform Tests (WPT),
and YTS Playback tests into the GitHub CI workflow. This ensures that a
wider range of functionality is automatically tested on pull requests
and pushes, improving stability and catching regressions earlier.

The CI configuration for relevant platforms is updated with new test
flags and targets. The main GitHub Actions workflow now includes new
jobs to execute these tests. Additionally, the on-device tests gateway
client is extended to recognize and process these new test types.

Bug: 473909877

---

💡 Pro Tips for a Better Commit Message:

1. Influence the Result: Want to change the output? You can write custom prompts or instructions directly in the Pull Request description. The model uses that text to generate the message.
2. Re-run the Generator: Post a comment with: /generate-commit-message

[gemini-code-assist]

Code Review

This pull request integrates Cobalt Browser tests and YTS WPT tests into the GitHub CI workflow by updating the configuration and the test gateway client script. The changes look good overall, but I've identified two potential issues. First, the configuration enables yts_playback tests, but the client script doesn't handle this test type, which will cause an error. Second, the handling for the new browser_test and yts_wpt_test types in the Python script is incomplete and duplicates code, which will likely cause those tests to fail. I've left detailed comments with suggestions on how to address these points by adding support for the new test type and refactoring the duplicated logic.

[gemini-code-assist]

This new block for browser_test and yts_wpt_test is missing logic that is present in the e2e_test/yts_test handler. Specifically, it's missing the addition of test_target to test_args, and appending test_root_target, system_image_download_url, and cobalt_path to files. This will likely cause these new tests to fail.

Instead of duplicating the logic, I recommend refactoring to combine this with the preceding elif block. You can modify the condition at line 248 to include the new test types and then add a nested if to handle the test_type reassignment. This will fix the bug and improve maintainability by removing code duplication.

Here's a suggested implementation:

1. Change line 248 to: elif test_type in ('e2e_test', 'yts_test', 'browser_test', 'yts_wpt_test'):
2. Add the following lines after line 248:

   if test_type in ('browser_test', 'yts_wpt_test'):
     test_type = 'e2e_test'

3. Remove this elif block (lines 264-271).

[Awallky]

Don't we want this on arm64 targets as well?
What about linux-x64x11?