Relay metrics

[viktorvsk]

Since knowstr already connects to relays, various metrics could be gathered, like:

• Connection latency
• Pink latency
• Whether relay sends/responds with ping
• Read/Write (different nostr commands) latency
• ...

However, it is in theory possible to have those numbers incorrect due to knowstr performance. For instance, if we want to sync 5000 relays with 1 worker, then high chances stats will be flawed not because of relays issues but because of knowstr overload