Syncer V2

[Kourin1996]

Description

Implementation of the new syncing protocol (Lightning Sync)

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)

Breaking changes

Yes this is a braking change

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

We have done a regression test for this module
Some of the test cases:

1. Syncing new node and continuing in watchSync state
2. Syncing multiple nodes at the same time with 1 validator
3. Stopping the network while in bulkSync and restarting to see will it continue
4. From watchSync to validator to watchSync

Documentation update

Will link when i open a PR

Additional comments

Fixes EDGE-592, EDGE-593, EDGE-594, EDGE-595

[zivkovicmilos]

I really don't like that you've ignored parts of the spec that we've agreed upon. We dropped some functionality and flows for a reason, but you chose to ignore that and reintroduced them back (pubsub is completely missing, Notify and Broadcast are back, no agreed upon error handling...).

Why are you supporting forking? It complicates the code exponentially, and makes it harder to read and work with, especially implement WatchSync that @Aleksao998 is working on. I especially don't like the complexity of the nofork package - I strongly believe we should drop it completely.

I don't see us introducing a consensus mechanism that supports forking in the near future, or even at all at this point - we can always add that functionality when we need it in a year or more, or ever.

There are also 0 tests in this PR, which is a shame since the whole point of us basing Lighting Sync on gRPC streams and RPC method calls, and simplifying the flow, was that it made it much easier to test and verify.