[syncer] Forgetting non-Dogechain protocol peer

[DarianShawn]

Description

Currently, peers will discover and connect to non-Dogechain peers, which is frustrating and extremely draining of connection resources.

The PR fixes it by forgetting (disconnect and remove it from store) the peer which is not supporting Dogechain syncer protocol.

Also, we print out the node ID that is currently syncing in eth_syncing endpoint and update its progress when the node status is updated.

Changes include

• Bugfix (non-breaking change that solves an issue)
• New feature (non-breaking change that adds functionality)

Testing

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

Manual tests

• Start a new MainNet node from Genesis block.
• Observe its peers protocol list.
  • dogechain peers list | grep "=" | awk '{print $3}' | while read p; do dogechain peers status --peer-id $p; done
• Observe its log.
• Observe its syncing status.
  • curl -X POST http://localhost:8545 -d '{"id": 0, "jsonrpc":"2.0", "method": "eth_syncing", "params": []}'

PR branch result:

• The connected nodes all support the Dogechain syncer protocol.
• There are some forget peer in the log.
• Return correct and upgraded progression.

Base branch result:

• Due to unfiltered protocol reply discovery, connected peers contain some non-Dogechain peers.

• No and never "forget" any peer in the log.

• Print out unobtrusive and sometimes erroneous sync status.

• The connected peers contain some non-Dogechain peers due to unfiltered protocol reply discovery.

• No and never 'forget' any peer in the log.

• Progress sometimes go wrong.

Documentation update

Should update official documentation once the version bumped.

[abrahamcruise321]

LGTM.