Fix Incorrect Metrics Caused by Applying Filters After Aggregation

Problem: Fix Incorrect Metrics Caused by Applying Filters After Aggregation You are working on an existing library management system. The application has a metrics page showing aggregate values, such as: Number of borrowed items Number of available items Number of overdue items Aggregated statistics by branch or category The metrics page supports filters such as: branch category status date range Users report that one value on the metrics page is incorrect after filters are selected. Known Information The frontend is written in JavaScript. The backend is written in Python using FastAPI. The actual issue is in the backend. The bug is that the backend computes metrics before applying filters, then filters the data afterward. Therefore the returned metric is computed from unfiltered data. Requirements Fix the backend logic so metrics are always computed from the filtered dataset. Constraints Do not change the API response shape. Preserve the existing filtering semantics. Add or fix tests covering: Correct metrics with no filters. Correct metrics with one filter. Correct metrics with multiple combined filters. Correct metrics when the filtered result is empty. Example Input Items: 3 available, 2 borrowed; no filters Output Metrics report available=3, borrowed=2.