fix(clv): Improve validation for missing columns in CLVModel (#1851)

* fix(clv): Improve validation for missing columns in CLVModel

Improves the column validation check to find all missing columns at
once and raises a single, informative ValueError listing all of them.

This directly addresses the goal of issue #1734 by providing a
much better user experience when data is incomplete.

* style: Apply pre-commit fixes

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: aaron1-z

pymc_marketing/clv/models/basic.py

@@ -75,20 +75,22 @@ def _validate_cols(
         must_be_unique: Sequence[str] = (),
         must_be_homogenous: Sequence[str] = (),
     ):
-        existing_columns = set(data.columns)
+        missing = set(required_cols).difference(data.columns)
+        if missing:
+            raise ValueError(
+                "The following required columns are missing from the "
+                f"input data: {sorted(list(missing))}"
+            )
+
         n = data.shape[0]
 
-        for required_col in required_cols:
-            if required_col not in existing_columns:
-                raise ValueError(f"Required column {required_col} missing")
-            if required_col in must_be_unique:
-                if data[required_col].nunique() != n:
-                    raise ValueError(f"Column {required_col} has duplicate entries")
-            if required_col in must_be_homogenous:
-                if data[required_col].nunique() != 1:
-                    raise ValueError(
-                        f"Column {required_col} has  non-homogeneous entries"
-                    )
+        for col in required_cols:
+            if col in must_be_unique:
+                if data[col].nunique() != n:
+                    raise ValueError(f"Column {col} has duplicate entries")
+            if col in must_be_homogenous:
+                if data[col].nunique() != 1:
+                    raise ValueError(f"Column {col} has non-homogeneous entries")
 
     def __repr__(self) -> str:
         """Representation of the model."""

tests/clv/models/test_basic.py

@@ -274,3 +274,17 @@ def test_deprecation_warning_on_old_config(self):
             model = CLVModelTest(model_config=old_model_config)
 
         assert model.model_config == {"x": Prior("Normal", mu=0, sigma=1)}
+
+    def test_validate_cols_reports_all_missing_columns(self):
+        """Test _validate_cols raises a single ValueError listing all missing columns."""
+        required = ("customer_id", "frequency", "recency", "T")
+        data = pd.DataFrame(
+            {
+                "customer_id": [1, 2, 3],
+                "frequency": [1, 2, 3],
+            }
+        )
+        expected_error_msg = "The following required columns are missing from the input data: ['T', 'recency']"
+
+        with pytest.raises(ValueError, match=expected_error_msg):
+            CLVModel._validate_cols(data=data, required_cols=required)