Enable Per-Refresh Memory Usage Monitoring in On-premises Data Gateway

Currently, Gateway Performance Logs provide only gateway-level memory and CPU metrics. There is no visibility into how much memory a specific semantic model refresh consumes

Please add per-refresh resource utilization metrics, including memory and CPU usage, so administrators can identify resource intensive refreshes, optimize schedules, prevent gateway resource exhaustion.