SIENTIAPDE-1255: Refactor MLFlow activities for training operations and update metrics

This commit refactors the MLFlow activities to focus on model training rather than prediction operations. It removes prediction-related activities and metrics, and updates the MLFlow activity descriptions to reflect the change in focus. The README is also updated to reflect these changes.

README.md

@@ -40,8 +40,7 @@ An enterprise-grade ML model training orchestration platform built on Temporal.
   - [Test Execution](#test-execution)
 - [Monitoring and Metrics](#monitoring-and-metrics)
   - [Application Health Metrics](#application-health-metrics)
-  - [Prediction Operation Metrics](#prediction-operation-metrics)
-  - [Data Quality Metrics](#data-quality-metrics)
+  - [Training Metrics](#training-metrics)
 - [Configuration](#configuration-1)
   - [Environment Variables](#environment-variables)
   - [Workflow Configuration](#workflow-configuration)
@@ -806,15 +805,6 @@ The Model Manager system exposes comprehensive Prometheus metrics for operationa
 - `app_up`: Application health status (1=healthy, 0=unhealthy)
   - Labels: `pod_id`
 
-### Prediction Operation Metrics
-- `model_manager_predictions_written_count`: Counter for successful prediction exports
-  - Labels: `pod_id`, `model_name`, `pipeline_name`
-- `model_manager_prediction_confidence_monitor`: Gauge for current prediction confidence levels
-  - Labels: `pod_id`, `model_name`, `pipeline_name`
-- `model_manager_prediction_response_time_monitor`: Histogram for prediction response times
-  - Labels: `pod_id`, `model_name`, `pipeline_name`
-  - Buckets: [0.01, 0.05, 0.1, 0.2, 0.5, 1.0, 2.0, 5.0, 10.0]
-
 ### Training Metrics
 - Training success/failure rates through notification system
 - Model save performance metrics