Align to spec v0.11.2

[terencechain]

🚀 Feature Request

Description

With the release of v0.11.2, this issue tracks to the progress to align Prysm to v0.11.2 which is important for multi client test net. This release does not contain breaking changes to the existing testnet

Beacon chain

• Remove redundant check in is_valid_indexed_attestation [non-substantive] (Fix #1735—remove redundant check ethereum/consensus-specs#1738)
• Clarify that state transitions with uint64 under- and over-flows are invalid (Partial fix for #1701 ethereum/consensus-specs#1739)

Networking

• Tighten restriction on "seen" attestation (Tighten restriction on a "seen" attestation ethereum/consensus-specs#1694)
• Notes on peer discovery/ENRs prior to genesis (Convert transaction descriptions from English to python code ethereum/consensus-specs#708, add note about distributing bootnode ENRs prior to genesis ethereum/consensus-specs#1730)
• Clarify attestation gossip condition and make "seen" cache per-epoch instead of per-slot (Redefine attestation propogation condition ethereum/consensus-specs#1706)
• Aggregate gossip condition "seen" cache per-epoch instead of per-slot (Tighten aggregate attn propogation condition ethereum/consensus-specs#1710)
• Clarify block range request description (simplify block range request description ethereum/consensus-specs#1736)
• Add missing note about "valid" attestation in gossip conditions (add missing word "valid" in attestation gossip conditions ethereum/consensus-specs#1744)
• Include fork digest in example gossip topics (Include fork digest in example gossip topic name ethereum/consensus-specs#1754)

[terencechain]

I'm gonna go against the spec and keep the is_valid_indexed_attestation indices length check. The spec assumes construction of the struct protects against invariant-length. Even if it does, it doesn't hurt to double check again on the application level

[terencechain]

This can be closed. Can you take a look at the networking change @nisdas before we close this? I think it's all done there's no action from our end

[nisdas]

There was actually one more change left to do, sorry I missed your comment @terencechain . I am opening a PR for it now