[Main2Main] Upgrade vllm commit to 0102

[wjunLu]

What this PR does / why we need it?

Upgrade vllm commit to 0102

1. Remove maybe_padded_num_tokens arg in model_runner_v1.py due to [Core] Remove unused num_tokens parameter from _init_model_kwargs vllm#31517
2. Remove Qwen/Qwen3-0.6B in tests/e2e/multicard/test_aclgraph_capture_replay.py because that Offline data parallel mode will be not supported/useful for dense models

Does this PR introduce any user-facing change?

How was this patch tested?

• vLLM version: v0.13.0
• vLLM main: vllm-project/vllm@7157596

[gemini-code-assist]

Code Review

This pull request updates the compatible vLLM commit hash for the main branch in the versioning policy documentation. The change is straightforward, but to improve long-term maintainability, I've suggested abstracting the hardcoded commit hash into a substitution variable. This aligns with the project's own documented policy for managing version-specific information in the documentation.

[gemini-code-assist]

For better maintainability, this commit hash should be defined as a substitution variable in docs/source/conf.py. This aligns with the project's documentation policy stated in this same file on line 138: 'To reduce maintenance costs, all branch documentation content should remain consistent, and version differences can be controlled via variables in docs/source/conf.py'.

You would need to add a new key to the myst_substitutions dictionary in docs/source/conf.py, for example:

myst_substitutions = {
    # ... existing substitutions
    'vllm_main_commit': 'ecd49ce7e69a50892be7f9841941ca2d7e3b12ea',
}

Then you can use the substitution here as suggested.

Suggested change

	| main | ecd49ce7e69a50892be7f9841941ca2d7e3b12ea, v0.13.0 tag | >= 3.10, < 3.12 | 8.3.RC2 | 2.8.0 / 2.8.0 |
	| main | {{ vllm_main_commit }}, v0.13.0 tag | >= 3.10, < 3.12 | 8.3.RC2 | 2.8.0 / 2.8.0 |

[jianzs]

Can we use GitHub Actions to automatically sync the latest commits in vllm?

[wjunLu]

Can we use GitHub Actions to automatically sync the latest commits in vllm?

We have this workflow https://github.com/vllm-project/vllm-ascend/actions/workflows/schedule_test_vllm_main.yaml to automatically sync and verfiy, but there are always some error so that we have to process them manually

[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.