GitLab

* reduce Routing Table churn

* peer diversity for the Routing Table.

* This is the fix for the bug wherein we don't fire PeerRemoved notifications when we replace "less useful" peers in a bucket.

`AddedAt` field for a peer in the Routing Table.

* reduce allocs while adding peer

* nPeerForCpl and collapse empty buckets

To determine how many buckets we should refresh, we:

1. Look at the max bucket.
2. Look at the peer with the greatest common prefix in the max bucket.
3. Return the "last refresh" times for all buckets between 0 and `min(maxPeer, 15)`

BREAKING: this returns a slice of times instead of CplRefresh objects because we
want to refresh _all_ buckets between 0 and the max CPL.

This reverts an optimization where we fudged the closer peers when returning
responses to users when we were within a power of two to the target.

* cache the XOR DHT ID for a peerId

* Routing Table refactor for new management logic

The neighbors may _also_ be empty or low. This can cause peers with a low number
of peers in their routing table to fail to find _any_ peers in their routing table.