add feature importances support for tuned models

[OkonSamuel]

This PR adds MLJModelInterface.feature_importances support for TunedModel.

Now calling MLJModelInterface.feature_importances on a TunedModel object just returns the feature importances for the wrapped model. For cases where the model wrapped by the TunedModel object doesn't support feature importance, nothing is returned.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 87.53%. Comparing base (c081bd0) to head (b9d8c01).
Report is 4 commits behind head on dev.

Additional details and impacted files

@@            Coverage Diff             @@
##              dev     #213      +/-   ##
==========================================
+ Coverage   87.42%   87.53%   +0.11%     
==========================================
  Files          13       13              
  Lines         660      666       +6     
==========================================
+ Hits          577      583       +6     
  Misses         83       83              

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

[ablaom]

Perfect. Thank you.