[Hotfix] Hotfix for paddlenlp_ops installing on sm90 for sage attention

[l1cacheDell]

Before submitting

• Lint code. If there are lint issues, please format the code first.

# Install and register `pre-commit` in the project folder
pip install pre-commit && pre-commit install

# Process previous code files separately
pre-commit run --file XXXX.py

• Add test cases into tests folder. If there are codecov issues, please add tests cases first.

PR types

PR changes

Description

[paddle-bot]

Thanks for your contribution!

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 48.67%. Comparing base (f0c5aa2) to head (b3a6f0e).
Report is 1 commits behind head on develop.

❌ Your project check has failed because the head coverage (48.67%) is below the target coverage (58.00%). You can increase the head coverage or adjust the target coverage.

Additional details and impacted files

@@           Coverage Diff            @@
##           develop   #10493   +/-   ##
========================================
  Coverage    48.67%   48.67%           
========================================
  Files          768      768           
  Lines       126915   126915           
========================================
+ Hits         61777    61779    +2     
+ Misses       65138    65136    -2     

☔ 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.

[vivienfanghuagood]

假如用户的/usr路径下安装了CUDA 12.3，但是pip安装了CUDA 12.6，会影响当前的判断吗？

[l1cacheDell]

假如用户的/usr路径下安装了CUDA 12.3，但是pip安装了CUDA 12.6，会影响当前的判断吗？

如果用户的/usr路径下安装了CUDA 12.3，就不会安装sage attention，不管paddle的版本是否是12.6。因为nvcc 12.3 编译sage attention会报错。