Revert onnxruntime-gpu version to 1.22.0 for Windows

[hthadicherla]

What does this PR do?

Type of change: ? Bug fix

Overview:
Reverted the windows ort version in setup.py to 1.22, since I later observed some regressions with vision models.
Keeping ORT version as 1.23 in windows examples since i didn't face any issues with LLM's.

Testing

Attaching results observed while testing these models.
int8_trt_comparison.csv

Summary by CodeRabbit

• Chores
  • Updated GPU runtime dependencies for the optional ONNX package. Simplified platform-specific version constraints while maintaining compatibility across supported platforms.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

📝 Walkthrough

Walkthrough

The setup.py file was modified to simplify the onnxruntime-gpu dependency specification in the "onnx" extras. A Windows-specific version constraint was removed, and the existing constraint was updated to remove the Windows platform exclusion, consolidating the configuration into a single rule.

Changes

Cohort / File(s)	Summary
Dependency Configuration setup.py	Simplified onnxruntime-gpu specification: removed Windows-specific override (==1.23.2) and consolidated to single constraint (\~=1.22.0) excluding aarch64 and Darwin platforms. Windows will now use the general constraint instead of a pinned version.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Title check	⚠️ Warning	The title 'Revert onnxruntime-gpu version to 1.22.0 for Windows' does not accurately reflect the actual change. The PR reverts the Windows constraint to use 1.22.0, but the change actually removes the Windows-specific constraint entirely, making 1.22.0 apply to non-Windows/non-Darwin platforms instead.	Update the title to reflect the actual change, such as 'Remove Windows-specific onnxruntime-gpu version constraint' or 'Update onnxruntime-gpu constraint to 1.22.0 for non-Darwin platforms' to accurately represent the modification.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 74.19%. Comparing base (615f99e) to head (350cf7b).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #801   +/-   ##
=======================================
  Coverage   74.19%   74.19%           
=======================================
  Files         192      192           
  Lines       19238    19238           
=======================================
  Hits        14273    14273           
  Misses       4965     4965           

☔ 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.