Run inference on the same device as used in the Trainer

[sokovninn]

This PR includes:

• Added inference logic to the LuxonisLightningModule
• Dataset and image folder inference now run on the same device as used in the Trainer
• Loss is no longer computed during inference

The implementation of video inference has not changed.

[kozlov721]

LGTM

[klemen1999]

LGTM

[codecov]

Codecov Report

Attention: Patch coverage is 97.61905% with 1 line in your changes missing coverage. Please review.

Project coverage is 95.79%. Comparing base (631b905) to head (bab2059).
Report is 8 commits behind head on main.

✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
luxonis_train/core/utils/infer_utils.py	97.22%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #121      +/-   ##
==========================================
- Coverage   96.31%   95.79%   -0.53%     
==========================================
  Files         147      159      +12     
  Lines        6304     6922     +618     
==========================================
+ Hits         6072     6631     +559     
- Misses        232      291      +59     

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