Change cultivated aggregation logic and make sensor choice an option

[emmaai]

Changes:

• the logic of cultivated changed from any sensor says cultivated -> cultivated to any sensor says natural -> natural as bias towards cultivated class was observed in validation
• sensor choice as an option for both cultivated and woody cover
• woody cover band name aligned with official definition woody instead of woody_cover

[tebadi]

Has the impact on the accuracy of validation data been measured? Not sure we have data from different sensors in the validation but if we do, it's worth checking this.

[codecov]

Codecov Report

Attention: Patch coverage is 57.57576% with 14 lines in your changes missing coverage. Please review.

Project coverage is 82.25%. Comparing base (182de0d) to head (729e294).
Report is 1 commits behind head on develop.

Files with missing lines	Patch %	Lines
odc/stats/plugins/lc_ml_treelite.py	41.66%	14 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #165      +/-   ##
===========================================
- Coverage    82.38%   82.25%   -0.14%     
===========================================
  Files           67       67              
  Lines         4951     4970      +19     
===========================================
+ Hits          4079     4088       +9     
- Misses         872      882      +10     

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

[emmaai]

Has the impact on the accuracy of validation data been measured? Not sure we have data from different sensors in the validation but if we do, it's worth checking this.

I have to merge this to be able to run another continental scale and then on validation data. There would be retrace potentially.