Fix sync committee when syncing

[tbenr]

fixes #8005

The aggregation fix is non-controversial while the sync message is not that clean, but it is the easiest implementation I came up with so far.

Essentially we will stop producing sync committee message if the head we have is older than 32 slot, which means:
1- we are connected to a BN which is falling out of sync.
2- the chain is not production a block for more than entire epoch, we have bigger problems.

Documentation

• I thought about documentation and added the doc-change-required label to this PR if updates are required.

Changelog

• I thought about adding a changelog entry, and added one if I deemed necessary.

[StefanBratanov]

LGTM

[tbenr]

@rolfyone I'm not a fan of the solution for the sync committee. The aggregate is fine.

dumping some messages wrote in chat:

essentially VC tracks the head, creates the message, and publish to the BN

so this is the only duty that is not created by the BN
so the creation can not be directly associated to a syncing state
to do a similar thing on VC we should have a ChainHeadTracker that becomes aware of the syncing state of the BN we are currently connected to

we could subscribe to sync event stream
but i think there are a lot of edge cases in case of failovers

[rolfyone]

so this is the only duty that is not created by the BN
so the creation can not be directly associated to a syncing state
to do a similar thing on VC we should have a ChainHeadTracker that becomes aware of the syncing state of the BN we are currently connected to

ultimately its 1 persistent duty for sync committees which is different to other dutes.. that's why the VC is doing more... I think it's practical overall, but it's definitely different to all the others...

[tbenr]

ultimately its 1 persistent duty for sync committees which is different to other dutes.. that's why the VC is doing more... I think it's practical overall, but it's definitely different to all the others...

I think It is fine yes. We could be wrong only if the chain is not producing blocks for an entire epoch. In that case i don't think there would be enough block space to include them all.