[P/D] [Misc] Remove useless PD check in deepseek

[liziyu179]

What this PR does / why we need it?

Remove useless PD check in deepseek

• vLLM version: v0.10.1.1
• vLLM main: vllm-project/vllm@6c7af81

[gemini-code-assist]

Code Review

This pull request removes hardcoded logic related to prefill/decode disaggregation, where the with_prefill flag was forced to specific values based on whether a node was a kv_producer or kv_consumer. While this simplifies the code by relying on the caller-provided context, it may introduce issues. Specifically, the changes could lead to incorrect behavior during profiling and graph capture for producer and consumer nodes in a disaggregated setup, as the assumptions about their roles during dummy runs seem to be violated. I've added two comments highlighting these potential issues.

[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.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 72.99%. Comparing base (984bd7c) to head (1b1da92).
⚠️ Report is 12 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2739      +/-   ##
==========================================
+ Coverage   72.57%   72.99%   +0.42%     
==========================================
  Files         150      153       +3     
  Lines       21372    21331      -41     
==========================================
+ Hits        15511    15571      +60     
+ Misses       5861     5760     -101     

Flag	Coverage Δ
unittests	72.99% <100.00%> (+0.42%)	⬆️

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.