feat(consensus): add the max_request_blocks field in ChainSpec

[duguorong009]

Issue Addressed

• Close Compatibility with Teku VC for voluntary-exit #4711

Proposed Changes

• Update the ChainSpec struct of consensus crate
  Add the max_request_blocks field

Additional Info

Originally, this PR is for addressing the issue of missing the MAX_REQUEST_BLOCKS info from lighthouse beacon API /eth/v1/config/spec.
While resolving the issue, I found that we don't need to update anything from BN crate code, but update the ChainSpec from consensus crate.
Hence, the PR is for updating the consensus crate, adding one more field to ChainSpec struct and related codes.

[michaelsproul]

Sorry, it looks like this problem is a bit more involved than described on the issue. We not only need to include the max_request_blocks in the chain spec, but also need to use this value in place of the existing MAX_REQUEST_BLOCKS constant. This will require refactoring some of the networking code, and should probably happen as part of the Deneb work (to avoid merge conflicts, and because Deneb also touches this code). I'll close the original issue and open a new one.

[divagant-martian]

There is #4562 @michaelsproul

[michaelsproul]

Ah! Thanks @divagant-martian, didn't see that!

[duguorong009]

@michaelsproul
So, what do you want me to do?
As far as I see, this PR is for Deneb work.
So, simply change the merge destination branch to Deneb one?

[divagant-martian]

HI @duguorong009 bare in mind that I'm not a team member but I know about this to some depth. What @michaelsproul is describing is that the changes you made address the issue partially and we want to address it fully. "Using" this value is what #4562 describes. This has not been done because in terms of code, it's not clear for us what's the best way to do it. Or at least it's definitely not straightforward.

This being said, re-targeting to deneb is not going to be enough, sadly

[duguorong009]

Thanks for the quick answer! @divagant-martian
It seems like this PR needs to wait for the discussion result. (currently, leaving as is)

If you and @michaelsproul find a wayout, and think this PR can continue, please let me know.

[michaelsproul]

Thanks for your work on this @duguorong009, it helped us clarify the issue. I'm going to close your PR now in favour of the more involved fix here: #4841.