eth-monitor all changes applied to node-finder/case-study so far

[SimonSK]

@zzma I'm concerned about the current re-dialing approach.

• For node-finder, I used a time ticker to check all known nodes' redial status every x seconds (default 5) rather than upon every loop because I was worried about the delay caused by iterating through the entire known node list all the time. See https://github.com/teamnsrg/go-ethereum/blob/cf274c2f4380658f0cc4a52e7366a5719398fdb0/p2p/server.go#L697-L729
• For eth-monitor (this PR), I removed the time ticker and left both queuing and starting tasks to occur every loop. See https://github.com/teamnsrg/go-ethereum/blob/cbbd3709ed8b95f6a5de5193b1f4fb0284cabfdd/p2p/server.go#L612-L638

What's your thought?