fix(network): expire stale and unreachable peers from the network participant list

[lukyrys]

The network participant list (Settings → LISH network → Participants) kept showing peers that went offline long ago, and the backend kept re-dialing them forever. This makes those entries expire and stops the churn.

• discovered peers with no live connection expire from the list 30 minutes after their last update, scoped per network (a peer still connected via another network no longer lingers under one it left)
• peers that stay unreachable across repeated re-dials, or have no reachable address, are evicted from the peer store and bootstrap/gossipsub state, then briefly quarantined so gossip can't instantly re-add them
• configured bootstrap peers are never evicted — a broken entry keeps its status row so it stays fixable
• an identity mismatch on one address of an otherwise-connected peer now trims just that address instead of dropping the whole peer; discovered rows are capped per network and only dial-verified addresses enter the peer store
• no frontend changes