using a tracker swarm module for redundancy and fall backs

[ducksandgoats]

this is a package i worked on, it creates a swarm of trackers using the bittorrent-dht package and if 1 tracker is busy or is overloaded with connections and resource usage, then that 1 tracker can send the connection to another tracker in the swarm.

it does this by sending a relay property along with the ['failure reason'] in returned object
if the client receives ['failure reason'] === 'Relaying', then the client can use the ['relay'] property, which is the announce url for another tracker in the swarm that the client can use since the original tracker is busy

https://github.com/ducksandgoats/bittorrent-relay

the tracker swarm creates redundancy, so i think it can be very complimentary to trystero

i made a fork of trystero to show what it looks like
https://github.com/ducksandgoats/trystero/blob/main/src/torrent.js#L91

here are two bittorrent-relay trackers i am running online
ws://retested.club:10509/announce and ws://retested.shop:10509/announce
also has http routes, try the http route /address.json it will show all the other trackers in the swarm in the form of host:port

just a small change to trystero since all we are doing is checking the ['failure reason'] to see if a ['relay'] property has been sent to us, what do you think about using this change?