Add sklearn estimators feature_importances_ attribute

TODO:
- check whether feature_evaluated_importances_ and
  feature_used_importances_ should be kept or not
- if not, then rename n_features_used_ to n_features_ and
  feature_used_names_ to feature_names_.
- accordingly, update tests and sklearn samples to reflect these changes.

Author: popescu-v

khiops/sklearn/estimators.py

@@ -1817,8 +1817,13 @@ class KhiopsClassifier(ClassifierMixin, KhiopsPredictor):
           to all features selected by the classifier. It ranges between 0 (little
           contribution to the model) and 1 (large contribution to the model).
 
-        - Importance: The geometric mean between the Level and the Weight.
+        - Importance: Average of the exact Shapley values of each used feature
+          across the training data.
 
+    feature_importances_ :  `ndarray <numpy.ndarray>` of shape (n_features_used\_, )
+        Importances of the features used by the classifier. The importance of each
+        feature is calculated as the average of its exact Shapley values across
+        the training data.
     is_multitable_model_ : bool
         ``True`` if the model was fitted on a multi-table dataset.
     model_ : `.DictionaryDomain`
@@ -2037,6 +2042,7 @@ def _fit_training_post_process(self, ds):
             )
             self.feature_used_names_ = feature_used_names_
             self.feature_used_importances_ = feature_used_importances_
+            self.feature_importances_ = self.feature_used_importances_[:, 2]
             self.n_features_used_ = len(self.feature_used_names_)
 
     def predict(self, X):