[FIX] Bootnode re-discovery issue

[ZeljkoBenovic]

Description

Right now, in a running network, once the bootnodes lose all of its peers (server restart, network issues or similar) it does not get back any peers, without restarting all the nodes.
There is a go routine that is supposed to ping random bootnode for the list of its peers, but the bootnodes stay disconnected from the rest of the network.

One can test this very easily, run a 4-node cluster with 1 bootnode. Keep all the nodes running but restart the bootnode. Once the bootnode goes back online, it will never join the network and it will always have 0 peers.

Changes include

• Bugfix (non-breaking change that solves an issue)
• Hotfix (change that solves an urgent issue, and requires immediate attention)
• New feature (non-breaking change that adds functionality)
• Breaking change (change that is not backwards-compatible and/or changes current functionality)

Checklist

• I have assigned this PR to myself
• I have added at least 1 reviewer
• I have added the relevant labels
• I have updated the official documentation
• I have added sufficient documentation in code

Testing

• I have tested this code with the official test suite
• I have tested this code manually

Manual tests

Run a 4-node cluster with 1 bootnode. Restart or power cycle the bootnode only.
After the bootnode comes back online, after a while (no more than 60s), it should start peering with all the other nodes.

Additional comments

Fixes EDGE-790

[epikichi]

Great catch. Confirmed locally that peers will dial back after bootnode(s) disruption occurs.

[Kourin1996]

LGTM. Thank you for correcting minor things in addition to the fixes

[0xAleksaOpacic]

LGTM, tested manually everything and the issue is fixed with this

[zivkovicmilos]

Looks good 💯

Thank you for hunting down this bug 🙏

I've left a few small comments that need to be addressed.

Please also resolve the conflicts in the go.mod file 🙏