[E2E][MM] Add e2e tests for InternVL model

[gcanlin]

What this PR does / why we need it?

As a validation for #3664, add end-to-end tests to monitor the InternVL model and ensure its continuous proper operation. This PR is only for single-card. So the models that have more parameters than 8B like 78B are needed to test using multi-cards.

Does this PR introduce any user-facing change?

None.

How was this patch tested?

pytest -sv tests/e2e/singlecard/multi-modal/test_internvl.py

• vLLM version: v0.11.0
• vLLM main: vllm-project/vllm@83f478b

[github-actions]

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

• A PR should do only one thing, smaller PRs enable faster reviews.
• Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
• Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

[gemini-code-assist]

Code Review

This pull request adds end-to-end tests for the InternVL model family, which is a good validation for the related changes. The test covers several models and correctly verifies the consistency between eager and graph execution modes. I've found a significant amount of duplicated code in the test logic for the two modes. I've suggested a refactoring to improve maintainability by removing this duplication.

[gcanlin]

@wangxiyuan InternVL2-8B and InternVL2_5-8B still have 27,985 and 8,270 downloads last month in huggingface. I think we should keep them in the e2e tests.
@MengqingCao The outputs of InternVL3_5-8B with eager mode and ACL graph mode equal exactly now. Don't know the reason lead to different and can't reproduce it. So remove check_outputs_similar now.
Please take a look again. Thanks!