Remove redundant and basic fit tests from unit suite

Deleted test_fitting_methods.py and test_likelihood_data_selection.py, and removed basic fit tests from test_classifier.py and test_decoder.py. These tests were either redundant, covered by integration tests, or too basic to provide meaningful coverage. This streamlines the test suite and focuses on more robust and relevant test cases.

Author: edeno

replay_trajectory_classification/tests/unit/test_classifier.py

@@ -123,26 +123,6 @@ def test_clusterless_classifier_construction():
     assert len(classifier.environments) == 2
 
 
-def test_clusterless_classifier_fit_basic():
-    """Test that ClusterlessClassifier can fit to data."""
-    environments = [make_1d_env("env1", n=10)]  # Use more bins
-    classifier = ClusterlessClassifier(environments=environments)
-
-    # Create simple training data with proper format
-    multiunit_data = make_simple_multiunit_data(n_electrodes=2, n_time=100)
-    position = np.linspace(0, 9, 100).reshape(-1, 1)  # Proper range
-
-    try:
-        classifier.fit(multiunit_data, position)
-        # Basic check that fitting completed without error
-        assert True  # If we get here, fit didn't crash
-    except Exception as e:
-        # If there's an environment/morphology issue, skip gracefully
-        if "tuple index out of range" in str(e) or "IndexError" in str(e):
-            pytest.skip(f"Environment fitting issue: {e}")
-        else:
-            raise
-
 
 def test_clusterless_classifier_predict_requires_fit():
     """Test that predict requires the classifier to be fit first."""

replay_trajectory_classification/tests/unit/test_decoder.py

@@ -154,25 +154,6 @@ def test_clusterless_decoder_construction():
     assert decoder.environment.environment_name == "test_env"
 
 
-def test_clusterless_decoder_fit_basic():
-    """Test that ClusterlessDecoder can fit to data."""
-    from replay_trajectory_classification.environments import Environment
-    environment = Environment(environment_name="test_env")
-    decoder = ClusterlessDecoder(environment=environment, infer_track_interior=False)
-
-    # Create simple training data with proper format and range
-    multiunit_data = make_simple_multiunit_data(n_electrodes=2, n_features=4, n_time=200)
-    position = np.linspace(0, 100, 200).reshape(-1, 1)  # Realistic track length
-
-    try:
-        decoder.fit(multiunit_data, position)
-        # Basic check that fitting completed without error
-        assert True  # If we get here, fit didn't crash
-    except IndexError as e:
-        if "tuple index out of range" in str(e):
-            pytest.skip(f"Environment boundary calculation issue with synthetic 1D data: {e}")
-        else:
-            raise
 
 
 def test_clusterless_decoder_predict_requires_fit():