Small edits to training

[sfmig]

Small edits to training to better match evaluation.

• factors out logging to screen of MLflow info and datasets info
• fixes typo in Detector class name
• CLI arguments review: reorder them to be more consistent across evaluation and training

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 71.42857% with 8 lines in your changes missing coverage. Please review.

Project coverage is 48.00%. Comparing base (c992a79) to head (c9070dc).

Files with missing lines	Patch %	Lines
crabs/detector/train_model.py	58.33%	5 Missing ⚠️
crabs/detector/evaluate_model.py	0.00%	3 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #235      +/-   ##
==========================================
+ Coverage   46.77%   48.00%   +1.23%     
==========================================
  Files          24       24              
  Lines        1520     1531      +11     
==========================================
+ Hits          711      735      +24     
+ Misses        809      796      -13     

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