Channel monitor watches for errors instead of measuring data rate

[dirkmc]

Depends on ipfs/go-graphsync#164
Supercedes #177 #172 #182 #181

Currently the channel monitor watches the data-rate and if it falls too low, the monitor sends a Restart message to the other party in the data transfer.

This is complex to do for retrieval deals, because the monitor needs to be aware of pauses and resumes.

Instead this PR simplifies the monitor's behaviour so that now:

• The monitor doesn't watch the data-rate
• The monitor just watches for errors fired from the transport layer
• When there is an error, the monitor tries to reconnect to the peer
• Once the connection is established, the monitor sends a restart message
• If there is no response to the restart message, the data transfer fails

[codecov-io]

Codecov Report

Merging #190 (689c5fe) into master (dbc57eb) will decrease coverage by 0.46%.
The diff coverage is 72.26%.

@@            Coverage Diff             @@
##           master     #190      +/-   ##
==========================================
- Coverage   65.42%   64.95%   -0.47%     
==========================================
  Files          25       25              
  Lines        2600     2594       -6     
==========================================
- Hits         1701     1685      -16     
- Misses        558      568      +10     
  Partials      341      341              

Impacted Files	Coverage Δ
channels/channels.go	70.88% <0.00%> (-0.46%)	⬇️
channels/channels_fsm.go	63.63% <0.00%> (-2.18%)	⬇️
network/libp2p_impl.go	68.90% <0.00%> (-2.40%)	⬇️
transport/graphsync/graphsync.go	76.19% <45.45%> (-0.42%)	⬇️
channelmonitor/channelmonitor.go	74.88% <81.05%> (-2.49%)	⬇️
impl/events.go	72.95% <100.00%> (+0.19%)	⬆️
impl/impl.go	59.55% <100.00%> (+1.07%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update dbc57eb...689c5fe. Read the comment docs.