Publish/validate before db commit

[paberr]

This PR adds a post-validation hook to the blockchain and uses it to publish/validate blocks before the database commit is happening. This way, the blocks are forwarded to the network without waiting for the database.

Pull request checklist

• All tests pass. The project builds and runs.
• I have resolved any merge conflicts.
• I have resolved all clippy and rustfmt warnings.

[nibhar]

Lots of additional Clone derives, which indicate that we now clone more. Not entirely sure where this is and how big the impact is, pretty sure it could be avoided though from what I read.

[nibhar]

In bad136e, which is on top of this PR, I removed a bunch of the clones again, and passed the arguments to post_validation by reference.

[nibhar]

In https://github.com/nimiq/core-rs-albatross/compare/nibhar/publish-pre-commit?expand=1 I also added the changed documentation, as well as a push_with_hook function which can be used to include the hook, whereas push remains unchanged. That cuts down on all those &() additions.

[Eligioo]

LGTM, nice way to decouple the publish and commit. 1 small log thing.

[styppo]

Looks like this event is now unused and we could get rid of it. But it's also not breaking anything so I think it would be fine to address this in a separate PR.

[nibhar]

Yes it is unused and could be removed.

Removing it will likely also entail making the micro Producer no longer having to implement Stream. That probably will mean a bunch of additional changes (for the better).

I opted not to do it here as it is out of scope of this PR.

[hrxi]

Maybe the "fix clippy warning" commit could be dropped instead of half-removing the MicroBlock event. A #[allow()] could be used instead, leaving the (full) removal for a future PR.

[nibhar]

For the time being I don't plan on doing any additional work there. The Event is unused, but removing the stream makes us less flexible in the future. If the micro producer is ever going to return something meaningful again it would have to be re implemented. It does not break anything as it is and I think it is semantically not too bad either.

[styppo]

LGTM

[jsdanielh]

Rebasing branch and signing commits for merge