Update chainspec p2p and etc

[0xmrree]

Issue Addressed

Missing values in /eth/v1/config/spec #8571
- there will be follow up PR for the re org props

Proposed Changes

1. As per above issue from EF dev ops, I added

"EPOCHS_PER_SUBNET_SUBSCRIPTION": "256",
  "ATTESTATION_SUBNET_COUNT": "64",
  "ATTESTATION_SUBNET_EXTRA_BITS": "0",
  "UPDATE_TIMEOUT": "8192",
  "DOMAIN_BLS_TO_EXECUTION_CHANGE": "0x0a000000"

to /eth/v1/config/spec
2. Had to change the minimal config for UPDATE_TIMEOUT to get currents tests to pass. This is ok given UPDATE_TIMEOUT is not used in lighthouse as this config for light client spec from altair
3. ATTESTATION_SUBNET_PREFIX_BITS is now dynamically calculated and shimmed into the /eth/v1/config/spec output as advised by @michaelsproul

Testing

1. ran cargo nextest run -p types and passed all tests, see below

...
        PASS [   0.006s] types withdrawal::withdrawal_request::tests::test_ssz_round_trip
        PASS [   0.007s] types withdrawal::withdrawal_request::tests::test_tree_hash_root
        PASS [   0.010s] types::committee_cache default_values
        PASS [   5.431s] types::committee_cache initializes_with_the_right_epoch
        PASS [   5.489s] types::committee_cache fails_without_validators
        PASS [   5.534s] types::state cache_initialization
        PASS [   5.617s] types::state beacon_proposer_index
        PASS [   0.238s] types::state decode_base_and_altair
        PASS [   5.740s] types::committee_cache min_randao_epoch_correct
        PASS [   6.236s] types::committee_cache shuffles_for_the_right_epoch
        PASS [  19.782s] types::state committees::current_epoch_committee_consistency
        PASS [  19.816s] types::state committees::next_epoch_committee_consistency
        PASS [  19.540s] types::state committees::previous_epoch_committee_consistency
────────────
     Summary [  25.263s] 265 tests run: 265 passed, 0 skipped

2. Started beacon node via ./target/release/lighthouse bn --network sepolia ... and ran

curl http://localhost:5052/eth/v1/config/spec | jq '.data | {
  EPOCHS_PER_SUBNET_SUBSCRIPTION,
  ATTESTATION_SUBNET_COUNT,
  ATTESTATION_SUBNET_EXTRA_BITS,
  ATTESTATION_SUBNET_PREFIX_BITS,
  UPDATE_TIMEOUT,
  DOMAIN_BLS_TO_EXECUTION_CHANGE
}'

which returned

{
  "EPOCHS_PER_SUBNET_SUBSCRIPTION": "256",
  "ATTESTATION_SUBNET_COUNT": "64",
  "ATTESTATION_SUBNET_EXTRA_BITS": "0",
  "ATTESTATION_SUBNET_PREFIX_BITS": "6",
  "UPDATE_TIMEOUT": "8192",
  "DOMAIN_BLS_TO_EXECUTION_CHANGE": "0x0a000000"
}

[michaelsproul]

You'll need to rebase on unstable to get CI to pass

[michaelsproul]

I'll be OOO until Jan 12 but someone might be able to review

[mergify]

Some required checks have failed. Could you please take a look @0xmrree? 🙏

[chong-he]

I have tested the beacon API output and the fields added in this PR are now displayed correctly as the output for mainnet.

Some comments below. Some CIs are failing too, so would be great if you can fix them, thanks.

[0xmrree]

@chong-he addressed the above feedback, ran cargo fmt --all, reran cargo nextest run -p types, and re tested api. the ci failures should be gone now that i have added default values for the added config fields. also the code is much simpler as i am not longer changing any values in the config anymore.

[chong-he]

This is looking much better, just a bit of minor comments and something that I ain't sure about - will need a dev look into this

Thanks!

[0xmrree]

addressed feedback and ready for another review

[chong-he]

addressed feedback and ready for another review

This now looks great to me, the beacon API query for /eth/v1/config/spec now contains the added 5 fields, thanks for the PR.

As this involves spec change, we will wait for a final review from the team in case I miss anything.

[0xmrree]

@chong-he if you have time it would be great if you could peek at #8613, should be really straight forward