[Fix] method always using cuda

[hiramf]

Motivation

single_level_grid_priors has a arg to specify device, which defaults to 'cuda'. this PR fixes the function to allow using cpu.

Modification

pass in the value of mask_feat.device to the method instead of converting the tensor afterwards.

BC-breaking (Optional)

no breaking changes

Use cases (Optional)

no new features

Checklist

1. Pre-commit or other linting tools are used to fix the potential lint issues.
2. The modification is covered by complete unit tests. If not, please add more unit tests to ensure the correctness.
3. If the modification has a dependency on downstream projects of a newer version, this PR should be tested with all supported versions of downstream projects.
4. The documentation has been modified accordingly, like docstring or example tutorials.

[CLAassistant]

All committers have signed the CLA.

[hiramf]

@RunningLeon would appreciate your review

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 48.38%. Comparing base (bc75c9d) to head (5f283be).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@ Coverage Diff @@ ## main #2715 +/- ## ======================================= Coverage 48.38% 48.38% ======================================= Files 357 357 Lines 13850 13850 Branches 2046 2046 ======================================= Hits 6702 6702 Misses 6693 6693 Partials 455 455 

Flag	Coverage Δ
unittests	48.38% <ø> (ø)

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.

[henryrobbins]

I ran into this issue and this patch fixed it for me as well!

[lauriebax]

If this MR doesn't get merged here, feel free to open it against our VBTI fork at https://github.com/VBTI-development/onedl-mmdeploy- we're actively maintaining updated versions with latest PyTorch support and would welcome the contribution!