[berkeley] Don't bootloop at startup for non-seed nodes

[mrmr1993]

This PR fixes a bootloop at startup for nodes without the --seed flag, where nodes will detect that they're offline (i.e. receiving no messages) and restart every 5 minutes.

A 'mitigation' for this previously existed, but didn't work as intended for the reasons given in the newly-added comment. This code should eventually be removed.

This is currently untested. I am travelling and have been so far unable to.

[mrmr1993]

Confirmed working as intended with the following command:

$ _build/default/src/app/cli/src/mina.exe daemon --config-file genesis_ledgers/mainnet.json --peer-list-url https://storage.googleapis.com/seed-lists/matt_test_seeds.txt
2024-05-06 18:24:07 UTC [Info] Mina daemon is booting up; built with commit "[DIRTY]f9e6c242f748a3aba3cfabc9362d70c8a38222e5" on branch "feature/no-bootloop-before-genesis"
2024-05-06 18:24:07 UTC [Info] Booting may take several seconds, please wait
2024-05-06 18:24:07 UTC [Info] Reading configuration files $config_files
        config_files: [ "/home/mathew/.mina-config/daemon.json", "genesis_ledgers/mainnet.json" ]
2024-05-06 18:24:07 UTC [Warn] Could not read configuration from "/home/mathew/.mina-config/daemon.json"
2024-05-06 18:24:07 UTC [Info] Initializing with runtime configuration. Ledger name: "mainnet"
2024-05-06 18:24:07 UTC [Info] Using the compiled constraint constants
2024-05-06 18:24:07 UTC [Info] Loaded genesis ledger from $ledger_file
        ledger_file: "/home/mathew/.mina-config/genesis/genesis_ledger_accounts_fe217061fe891753eec9838e1f4edc169ff6c7f28ad0e9628f2167e6716f5fec.tar.gz"
2024-05-06 18:24:07 UTC [Info] Cleaning up old epoch ledgers with genesis state $state_hash at locations $staking and $next
        state_hash: "3NKVWi3D9z3bJeG5pTbV32sZDSKVwAfaGi83ToX2pdHGCAudcoYq"
        staking: "/home/mathew/.mina-config/epoch_ledgerc2490d55-513e-3239-c227-b3e867c6bafc"
        next: "/home/mathew/.mina-config/epoch_ledger6634046c-1b6e-3e14-eb42-d5f2bd651bd8"
2024-05-06 18:24:25 UTC [Info] Daemon will use chain id 38ac2bb187ecc81307a4c23e914c202132f91dbb22360f340e15d312e1fddc97
2024-05-06 18:24:25 UTC [Info] Daemon running protocol version 3.0.0
2024-05-06 18:24:25 UTC [Info] Node started before genesis time "2024-05-06T18:35:30.000000Z". Sleeping for "3m"
2024-05-06 18:27:25 UTC [Info] Node started before genesis time "2024-05-06T18:35:30.000000Z". Sleeping for "3m"
2024-05-06 18:30:25 UTC [Info] Node started before genesis time "2024-05-06T18:35:30.000000Z". Sleeping for "2.083m"
2024-05-06 18:32:30 UTC [Info] Starting a new prover process
2024-05-06 18:32:49 UTC [Info] Daemon started process of kind "Prover" with pid 114141
2024-05-06 18:32:49 UTC [Info] Starting a new verifier process
2024-05-06 18:33:06 UTC [Info] Daemon started process of kind "Verifier" with pid 114168
2024-05-06 18:33:06 UTC [Info] Starting a new vrf-evaluator process
2024-05-06 18:33:06 UTC [Info] Daemon started process of kind "Vrf_evaluator" with pid 114229
Environment variable MINA_TIME_OFFSET not found, using default of 0
2024-05-06 18:33:07 UTC [Error] Invalid peer $peer found in peers list
        peer: "<?xml version='1.0' encoding='UTF-8'?><Error><Code>NoSuchKey</Code><Message>The specified key does not exist.</Message><Details>No such object: seed-lists/matt_test_seeds.txt</Details></Error>"
2024-05-06 18:33:07 UTC [Info] Custom child process "libp2p_helper" started with pid 114236
2024-05-06 18:33:07 UTC [Error] libp2p_helper: failed to set log level debug for routing/record: error: No such logger
2024-05-06 18:33:07 UTC [Info] libp2p peer ID this session is $peer_id
        peer_id: "12D3KooWKQTRme83KA1mGpCVsLW5hHYNdxY6pujCZpC2oLtW2tHd"
2024-05-06 18:33:07 UTC [Info] Network not instantiated when node status requested
2024-05-06 18:33:07 UTC [Info] Starting transition router
2024-05-06 18:33:07 UTC [Warn] Node started before the chain start time: waiting 142175 milliseconds before starting participation
2024-05-06 18:33:07 UTC [Warn] GCLOUD_KEYFILE environment variable not set. Must be set to use upload_blocks_to_gcloud
2024-05-06 18:33:07 UTC [Info] Mina daemon is connecting
2024-05-06 18:33:07 UTC [Info] Initializing plugins
2024-05-06 18:33:07 UTC [Info] Not running uptime service, no URL given
2024-05-06 18:33:07 UTC [Info] Stopping daemon after 10620 mins and when there are no blocks to be produced
2024-05-06 18:33:07 UTC [Info] Attempted to turn on snark worker, but snark worker key is set to none
2024-05-06 18:33:07 UTC [Info] Daemon ready. Clients can now connect
2024-05-06 18:33:07 UTC [Info] Created GraphQL server at: http://localhost:3085/graphql
2024-05-06 18:33:07 UTC [Error] libp2p_helper: failed to routing advertise: failed to find any peer in table
2024-05-06 18:33:07 UTC [Warn] starting libp2p up failed: $error
        error: {
  "commit_id": "[DIRTY]f9e6c242f748a3aba3cfabc9362d70c8a38222e5",
  "string": "libp2p error: failed to find any peer in table"
}

$ head -n 5 -- genesis_ledgers/mainnet.json
{
  "genesis": {
    "genesis_state_timestamp": "2024-05-06T18:35:30Z"
  },
  "ledger": {

[emberian]

This is horrifying and indeed explains the mysterious behavior i have been seeing