feat: enhance training and regression metrics logging

- Added debug logging for data preparation, transformation, and prediction steps in the Training class to improve traceability.
- Updated compute_regression_metrics method to include metadata for better debugging and validation of index alignment between true and predicted values.

model_manager/activities/training.py

@@ -13,6 +13,7 @@ with workflow.unsafe.imports_passed_through():
     from typing import Any
 
     import mlflow
+    import pandas as pd
     from sientia_do.notifications.handlers import CoreNotificationHandler as NotificationHandler
     from sientia_do.notifications.models import NotificationLevel
     from sientia_do.observability.logger import Logger
@@ -231,6 +232,12 @@ class Training(SientiaMonitoring):
             train_data = train_result.train_data
             val_data = train_result.val_data
 
+            self.debug(
+                f'train_model prepared data (head 10):\ntrain:\n{train_data.head(10).to_string()}'
+                f'\nval:\n{val_data.head(10).to_string()}',
+                metadata,
+            )
+
             wrapper.train(
                 train_data=train_data,
                 val_data=val_data,
@@ -245,9 +252,21 @@ class Training(SientiaMonitoring):
             transformed_train, _ = wrapper.transform(train_data)
             transformed_val, _ = wrapper.transform(val_data)
 
+            self.debug(
+                f'train_model transform (head 10):\ntrain:\n{transformed_train.head(10).to_string()}'
+                f'\nval:\n{transformed_val.head(10).to_string()}',
+                metadata,
+            )
+
             y_train_pred_df, _ = wrapper.predict({}, transformed_train)
             y_val_pred_df, _ = wrapper.predict({}, transformed_val)
 
+            self.debug(
+                f'train_model predict (head 10):\ntrain:\n{y_train_pred_df.head(10).to_string()}'
+                f'\nval:\n{y_val_pred_df.head(10).to_string()}',
+                metadata,
+            )
+
             y_train_pred_df.sort_index(inplace=True, ascending=False)
             y_val_pred_df.sort_index(inplace=True, ascending=False)
 
@@ -258,6 +277,7 @@ class Training(SientiaMonitoring):
             train_result = self.data_manager_repository.compute_regression_metrics(
                 train_result,
                 wrapper,
+                metadata=metadata,
             )
 
             self.info(f'Starting MLflow run for {train_params.model_type}', metadata)