Make confusion matrix work with evaluate module (#1444)

* make cm work with evaluate

* Update test_confusion.py

docs/releases/unreleased.md

@@ -39,6 +39,10 @@ River's mini-batch methods now support pandas v2. In particular, River conforms
 
 - Simplify inner the structures of `forest.ARFClassifier` and `forest.ARFRegressor` by removing redundant class hierarchy. Simplify how concept drift logging can be accessed in individual trees and in the forest as a whole.
 
+## metrics
+
+- `metrics.ConfusionMatrix` may now be used with `evaluate.progressive_val_score` and `evaluate.iter_progressive_val_score`.
+
 ## proba
 
 - Added `_from_state` method to `proba.MultivariateGaussian` to warm start from previous knowledge.

river/metrics/confusion.py

@@ -127,3 +127,10 @@ def total_false_positives(self):
     @property
     def total_false_negatives(self):
         return sum(self.false_negatives(label) for label in self.classes)
+
+    def works_with(self, model) -> bool:
+        return utils.inspect.isclassifier(model)
+
+    @property
+    def requires_labels(self):
+        return True

river/metrics/test_confusion.py

@@ -0,0 +1,16 @@
+from __future__ import annotations
+
+from river import datasets, evaluate, linear_model, metrics, optim, preprocessing
+
+
+def test_issue_1443():
+    dataset = datasets.Phishing()
+
+    model = preprocessing.StandardScaler() | linear_model.LogisticRegression(
+        optimizer=optim.SGD(0.1)
+    )
+
+    metric = metrics.ConfusionMatrix()
+
+    for _ in evaluate.iter_progressive_val_score(dataset, model, metric):
+        pass