[CI] Prune tests/models/decoder_only/language/* tests

[mgoin]

• Get rid of test_big_models.py to fold in different small versions of architectures into test_models.py
• Use small models for test_fp8.py
• Test less models in test_gptq_marlin.py, test_gptq_marlin_24.py
• Get rid of test_marlin.py and test_qwen.py
• Make test_mistral.py more efficient slightly

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.
Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can do one of these:

• Add ready label to the PR
• Enable auto-merge.

🚀

[DarkLight1337]

Thanks for saving CI costs!

[DarkLight1337]

It seems there are some failures in the tests though, please fix them.