restrict best LC update collection to canonical blocks

[etan-status]

Simplify best LightClientUpdate collection by tracking only canonical data instead of tracking the best update across all branches within the sync committee period.

• Restrict best LC update collection to canonical blocks ethereum/consensus-specs#3553

[github-actions]

Unit Test Results

         9 files  ±0    1 098 suites  ±0   27m 22s ⏱️ -12s
  3 951 tests ±0    3 604 ✔️ ±0  347 💤 ±0  0 ❌ ±0 
16 066 runs  ±0  15 668 ✔️ ±0  398 💤 ±0  0 ❌ ±0 

Results for commit cc2bb10. ± Comparison against base commit 375f276.

[jakubgs]

This PR looks like the most likely culprit for this crash which is present on all our unstable nodes on Holesky testnet:

../csu/libc-start.c(392)
../sysdeps/nptl/libc_start_call_main.h(58)
..../vendor/nim-libp2p/libp2p/protocols/rendezvous.nim(902) main
..../vendor/nim-libp2p/libp2p/protocols/rendezvous.nim(895) NimMain
..../beacon_chain/nimbus_beacon_node.nim(2346) main
..../beacon_chain/nimbus_beacon_node.nim(2214) handleStartUpCmd
..../beacon_chain/nimbus_beacon_node.nim(2102) doRunBeaconNode
..../beacon_chain/nimbus_beacon_node.nim(521) init
..../vendor/nim-chronos/chronos/internal/asyncfutures.nim(359) futureContinue
..../beacon_chain/nimbus_beacon_node.nim(717) init
..../beacon_chain/nimbus_beacon_node.nim(241) loadChainDag
..../beacon_chain/consensus_object_pools/blockchain_dag.nim(1131) init
..../vendor/nimbus-build-system/vendor/Nim/lib/system/assertions.nim(38) failedAssertImpl
..../vendor/nimbus-build-system/vendor/Nim/lib/system/assertions.nim(28) raiseAssert
..../vendor/nimbus-build-system/vendor/Nim/lib/system/fatal.nim(54) sysFatal
[[reraised from:
../csu/libc-start.c(392)
../sysdeps/nptl/libc_start_call_main.h(58)
..../vendor/nim-libp2p/libp2p/protocols/rendezvous.nim(902) main
..../vendor/nim-libp2p/libp2p/protocols/rendezvous.nim(895) NimMain
..../beacon_chain/nimbus_beacon_node.nim(2346) main
..../beacon_chain/nimbus_beacon_node.nim(2214) handleStartUpCmd
..../beacon_chain/nimbus_beacon_node.nim(2102) doRunBeaconNode
..../beacon_chain/nimbus_beacon_node.nim(521) init
..../vendor/nim-chronos/chronos/internal/asyncfutures.nim(359) futureContinue
..../beacon_chain/nimbus_beacon_node.nim(845) init
]]
Error: unhandled exception: ..../beacon_chain/consensus_object_pools/blockchain_dag.nim(1131, 12) `dag.finalizedHead.blck.parent == nil` ...but that's the last BlockRef with a parent [AssertionDefect]

[jakubgs]

Correction: I found a Prater node affected on: unstable-large-01.aws-eu-central-1a.nimbus.prater

[etan-status]

This PR looks like the most likely culprit for this crash

• fix startup after BN exited between head and finalized blocks updates #5617