[backport] Compatibility fixes for scikit-learn 1.6 (#11021, #11162)

[hcho3]

No description provided.

[hcho3]

Just tried the following snippet:

from sklearn.model_selection import GridSearchCV
from sklearn.datasets import load_breast_cancer
from xgboost import XGBClassifier

X, y = load_breast_cancer(return_X_y=True)

param_grid = {"learning_rate": [0.1, 0.2]}
xgb = XGBClassifier(objective='binary:logistic')
clf = GridSearchCV(xgb, param_grid, scoring='accuracy', cv=5, verbose=1)
clf.fit(X, y)

Before this patch:

$ python test.py
Traceback (most recent call last):
  File "/home/ubuntu/xgboost/test.py", line 10, in <module>
    clf.fit(X, y)
  File "/home/ubuntu/miniforge3/envs/foobar/lib/python3.10/site-packages/sklearn/base.py", line 1389, in wrapper
    return fit_method(estimator, *args, **kwargs)
  File "/home/ubuntu/miniforge3/envs/foobar/lib/python3.10/site-packages/sklearn/model_selection/_search.py", line 933, in fit
    cv_orig = check_cv(self.cv, y, classifier=is_classifier(estimator))
  File "/home/ubuntu/miniforge3/envs/foobar/lib/python3.10/site-packages/sklearn/base.py", line 1237, in is_classifier
    return get_tags(estimator).estimator_type == "classifier"
  File "/home/ubuntu/miniforge3/envs/foobar/lib/python3.10/site-packages/sklearn/utils/_tags.py", line 430, in get_tags
    sklearn_tags_provider[klass] = klass.__sklearn_tags__(estimator)  # type: ignore[attr-defined]
  File "/home/ubuntu/miniforge3/envs/foobar/lib/python3.10/site-packages/sklearn/base.py", line 540, in __sklearn_tags__
    tags = super().__sklearn_tags__()
AttributeError: 'super' object has no attribute '__sklearn_tags__'

After this patch:

$ python test.py
Fitting 5 folds for each of 2 candidates, totalling 10 fits

[jakirkham]

Huzzah! 🥳

Thanks Hyunsu 🙏