fix: clear input example in Training class before model storage

- Set wrapper._input_example to None to ensure no residual data is stored with the model, enhancing data integrity during model storage.

model_manager/activities/training.py

@@ -338,6 +338,7 @@ class Training(SientiaMonitoring):
             raise ValueError('Report path, train data path, or test data path is not set')
 
         self.info(f'Storing model for {train_params.model_type}', metadata)
+        wrapper._input_example = None
         wrapper.store_model(name=train_params.model_name)
         self.info(f'Logging artifacts for {train_params.model_type}', metadata)
         mlflow.log_artifact(train_result.report_path)