misc: Drop the raft-cluster experiment

[magik6k]

Related Issues

• This PR mostly reverts the raft cluster experiment from Raft consensus for lotus nodes in a cluster #9294
  • We keep the multi-node RPC feature, it is not perfect, but it is useful
    • I'll push some improvements to that feature in a follow-up PR

Proposed Changes

This PR removes raft cluster support entirely. This feature has never graduated beyond an experiment, had poor UX (e.g. no way to manage a running cluster, so it didn't even provide HA), and pulled in a lot of heavy dependencies. This implementation also is only useful in clusters which want to do message signing on chain nodes, and in most cases where running a node cluster is warranted, signing anything on nodes which are high maintenance chain syncing nodes is at least a bad practice.

Additional Info

Checklist

Before you mark the PR ready for review, please make sure that:

• Commits have a clear commit message.
• PR title is in the form of of <PR type>: <area>: <change being made>
  • example: fix: mempool: Introduce a cache for valid signatures
  • PR type: fix, feat, build, chore, ci, docs, perf, refactor, revert, style, test
  • area, e.g. api, chain, state, market, mempool, multisig, networking, paych, proving, sealing, wallet, deps
• If the PR affects users (e.g., new feature, bug fix, system requirements change), update the CHANGELOG.md and add details to the UNRELEASED section.
• New features have usage guidelines and / or documentation updates in
  • Lotus Documentation
  • Discussion Tutorials
• Tests exist for new functionality or change in behavior
• CI is green

[Stebalien]

TODO: Add this to the release notes.