[ENH] Bagging ensemble of probabilistic regressors

[fkiraly]

Adds bagging ensemble of probabilistic regressors.

Complete rewrite of the legacy estimator.

Depends on:

• [ENH] mixture of distributions #26 for the mixture distribution used in bagging.
• [ENH] enable default for predict in probabilistic regressors if only interval predictions are possible #65 for the default of predict, if _predict_proba or other proba methods are implemented but not _predict.

[fkiraly]

predict_quantiles takes long to run since ppf is computed by bisection, in the mixture. Should probably try to tweak the bisection parameters (higher tolerance, less max iterations etc) to ensure shorter runtime.

[codecov-commenter]

Codecov Report

Patch coverage: 100.00% and project coverage change: +0.65% 🎉

Comparison is base (0d8bd68) 70.91% compared to head (a8dc486) 71.57%.
Report is 1 commits behind head on main.

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #32      +/-   ##
==========================================
+ Coverage   70.91%   71.57%   +0.65%     
==========================================
  Files          93       93              
  Lines        4962     5016      +54     
  Branches      900      903       +3     
==========================================
+ Hits         3519     3590      +71     
+ Misses       1205     1189      -16     
+ Partials      238      237       -1     

Files Changed	Coverage Δ
skpro/regression/base/_base.py	65.40% <100.00%> (+1.07%)	⬆️
skpro/regression/ensemble.py	100.00% <100.00%> (+55.17%)	⬆️
skpro/regression/tests/test_all_regressors.py	100.00% <100.00%> (ø)

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