feat: Use rich Panel for showing warning in train_test_split (#1086)

closes #1060 

It is the alternative to #1060 using `rich`. I added a test to check
that we can filter the warning since we are not using the usual
`warnings` module.

In the future, we could factor out the code in a utils to be sure that
we can also transform the warnings into error.

Author: glemaitre

skore/src/skore/sklearn/train_test_split/train_test_split.py

@@ -7,6 +7,7 @@
 
 import numpy as np
 from numpy.random import RandomState
+from rich.panel import Panel
 
 from skore.sklearn.find_ml_task import _find_ml_task
 from skore.sklearn.train_test_split.warning import TRAIN_TEST_SPLIT_WARNINGS
@@ -88,26 +89,27 @@ class labels.
     >>> X, y = np.arange(10).reshape((5, 2)), range(5)
 
     >>> # Drop-in replacement for sklearn train_test_split
-    >>> X_train, X_test, y_train, y_test = train_test_split(X, y,
+    >>> X_train, X_test, y_train, y_test = train_test_split(X, y,  # doctest: +SKIP
     ...     test_size=0.33, random_state=42)
-    >>> X_train
+    >>> X_train  # doctest: +SKIP
     array([[4, 5],
            [0, 1],
            [6, 7]])
 
     >>> # Explicit X and y, makes detection of problems easier
-    >>> X_train, X_test, y_train, y_test = train_test_split(X=X, y=y,
+    >>> X_train, X_test, y_train, y_test = train_test_split(X=X, y=y,  # doctest: +SKIP
     ...     test_size=0.33, random_state=42)
-    >>> X_train
+    >>> X_train  # doctest: +SKIP
     array([[4, 5],
            [0, 1],
            [6, 7]])
 
     >>> # When passing X and y explicitly, X is returned before y
     >>> arr = np.arange(10).reshape((5, 2))
-    >>> arr_train, arr_test, X_train, X_test, y_train, y_test = train_test_split(
+    >>> splits = train_test_split(  # doctest: +SKIP
     ...     arr, y=y, X=X, test_size=0.33, random_state=42)
-    >>> X_train
+    >>> arr_train, arr_test, X_train, X_test, y_train, y_test = splits  # doctest: +SKIP
+    >>> X_train  # doctest: +SKIP
     array([[4, 5],
            [0, 1],
            [6, 7]])
@@ -158,10 +160,24 @@ class labels.
         ml_task=ml_task,
     )
 
+    from skore import console  # avoid circular import
+
     for warning_class in TRAIN_TEST_SPLIT_WARNINGS:
         warning = warning_class.check(**kwargs)
 
-        if warning is not None:
-            warnings.warn(message=warning, category=warning_class, stacklevel=1)
+        if warning is not None and (
+            not warnings.filters
+            or not any(
+                f[0] == "ignore" and f[2] == warning_class for f in warnings.filters
+            )
+        ):
+            console.print(
+                Panel(
+                    title=warning_class.__name__,
+                    renderable=warning,
+                    style="orange1",
+                    border_style="cyan",
+                )
+            )
 
     return output

skore/tests/unit/sklearn/train_test_split/test_train_test_split.py

@@ -129,15 +129,45 @@ def case_time_based_column_polars_dates():
         case_time_based_column_polars_dates,
     ],
 )
-def test_train_test_split_warns(params):
+def test_train_test_split_warns(params, capsys):
     """When train_test_split is called with these args and kwargs, the corresponding
-    warning should fire."""
-    warnings.simplefilter("ignore")
+    warning should be printed to the console."""
+    args, kwargs, warning_cls = params()
+
+    train_test_split(*args, **kwargs)
+
+    captured = capsys.readouterr()
+    assert warning_cls.__name__ in captured.out
+
+
+@pytest.mark.parametrize(
+    "params",
+    [
+        case_high_class_imbalance,
+        case_high_class_imbalance_too_few_examples,
+        case_high_class_imbalance_too_few_examples_kwargs,
+        case_high_class_imbalance_too_few_examples_kwargs_mixed,
+        case_stratify,
+        case_random_state_unset,
+        case_shuffle_true,
+        case_shuffle_none,
+        case_time_based_column,
+        case_time_based_columns_several,
+        case_time_based_column_polars,
+        case_time_based_column_polars_dates,
+    ],
+)
+def test_train_test_split_warns_suppressed(params, capsys):
+    """Verify that warnings can be suppressed and don't appear in the console output."""
     args, kwargs, warning_cls = params()
 
-    with pytest.warns(warning_cls):
+    with warnings.catch_warnings():
+        warnings.filterwarnings("ignore", category=warning_cls)
         train_test_split(*args, **kwargs)
 
+    captured = capsys.readouterr()
+    assert warning_cls.__name__ not in captured.out
+
 
 def test_train_test_split_kwargs():
     """Passing data by keyword arguments should produce the same results as passing