[6/N][refactor]delete torchair in rotary ops

[hust17yixuan]

What this PR does / why we need it?

After moved torchair related rope ops into torchair_ops, split the torchair from the origin rope ops to make the code clean.

Does this PR introduce any user-facing change?

No

How was this patch tested?

vLLM version: main
vLLM main: vllm-project/vllm@ab9f2cf

• vLLM version: v0.10.1.1
• vLLM main: vllm-project/vllm@81eea3d

[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 refactors the rotary embedding operations to remove torchair-specific logic, aiming for cleaner code. The changes correctly remove conditional blocks related to torchair from rope_forward_oot, AscendRotaryEmbedding.__init__, and AscendRotaryEmbedding.forward_oot. However, this refactoring has left some function parameters unused, which should be removed to complete the cleanup.

[gemini-code-assist]

As part of the torchair code removal, several parameters have become unused. To align with the PR's goal of cleaning up the code, it's best to remove these dead parameters to improve clarity and maintainability.

• The parameter is_qwen_torchair is passed here but is no longer used in rope_forward_oot after the refactoring. It should be removed from this call and from the function definition of rope_forward_oot.
• The parameters max_seq_len and is_prefill in the signature of forward_oot are also no longer used and should be removed.

Removing these unused parameters will make the code cleaner and prevent potential confusion for future developers.

        return rope_forward_oot(self, positions, query, key, offsets,
                                is_neox_style_override)  # type: ignore

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 73.06%. Comparing base (3a5fc5e) to head (4c2da15).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2581      +/-   ##
==========================================
+ Coverage   73.03%   73.06%   +0.03%     
==========================================
  Files         149      149              
  Lines       21515    21474      -41     
==========================================
- Hits        15714    15691      -23     
+ Misses       5801     5783      -18     

Flag	Coverage Δ
unittests	73.06% <100.00%> (+0.03%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.