Overhaul stats

[josecelano]

Depends on: #1319
Relates to: #1368

There are some inconsistencies in how we handle stats events and also some things that can be improved.

I'm also considering keeping metrics for each socket in addition to the aggregate metrics. We are currently collecting single aggregated metrics for for all servers (grouped by UDP or HTTP trackers).

UPDATE 2025-03-17

This refactor is needed to enrich events. The final goal is to expose those events to allow third-party consumers build their metrics. For example: you can create your event listeners to collects metrics for your private Index from Torrust Tracker.

cc @da2ce7

Notes

• The issue When I configured the bind_address to [::]:6969 to listen on both IPv4 and IPv6 addresses simultaneously, the tcp4 and udp4 in the HTTP API both showed 0 #1368 migth imply opening new sub-issues for this epic.