Centralize metrics in dedicated module

[Lore09]

This pull request introduces Prometheus metrics support to the project, enabling the collection and exposure of key operational and security metrics at a configurable endpoint. The implementation is configurable via config.yaml and can be toggled on or off. The changes also refactor metric-related code into a dedicated module and update background tasks to refresh metrics as part of their normal execution.

Prometheus metrics integration:

• Added a new metrics section to config.yaml and corresponding metrics_enabled configuration in Config, allowing Prometheus metrics to be enabled or disabled via configuration. [1] [2] [3] [4]

• Introduced a new src/metrics.py module that defines all Prometheus metric gauges and provides refresh helpers for updating metric values from background tasks.

Metrics endpoint and routing:

• Exposed a /metrics endpoint under the dashboard secret path, serving Prometheus-formatted metrics only when enabled in the config; otherwise, returns a 404. [1] [2]

Background task integration:

• Refactored background tasks (analyze_ips.py and dashboard_warmup.py) to use the new metrics refresh helpers, removing direct Prometheus usage from these files and ensuring metrics are updated as part of existing workflows. [1] [2] [3] [4] [5] [6]

Code cleanup:

• Removed the previous direct mounting of the Prometheus ASGI app from app.py, centralizing metrics routing in the dashboard router for better configurability and separation of concerns. [1] [2]