Add CLN ln node to the environment

[optout21]

Add CLN process to the devenv processes

• Stock CLN
• Taken from pkgs.clightning nix package
• devenv had to be updated, for CLN 25.0, as v 24.0 does not know testnet4
• connects to bitcoind

Partly fixes #50 .

[vnprc]

This is a great start! Thank you for working on this.

The cln node dies on startup when it fails to connect to the bitcoind node. I can think of two causes:

1. Double check the port. You may need to add a new entry to the config/miner.toml file. This is a global config file that keeps track of settings shared across roles. There's a miner.toml and a pool.toml to distinguish between the poolside services and the minerside services. Check this diagram: https://hashpool.dev/images/diagrams/process-architecture.png. Lightning is tightly integrated with the mint so it goes on the pool side.

2. If that doesn't get it working you may need to add a polling loop to force cln to wait for bitcoind to come up. See the polling loops in other processes for an example:

    pool = {
      exec = withLogging ''
        echo "Waiting for Mint..."
        while ! nc -z localhost ${toString poolConfig.mint.port}; do
          sleep 1
        done
        echo "Mint is up. Starting Local Pool..."
        cargo -C roles/pool -Z unstable-options run -- \
          -c ${config.devenv.root}/roles/pool/config-examples/pool-config-local-tp-example.toml \
          -g ${config.devenv.root}/config/pool.toml
      '' "pool.log";
    };

[optout21]

I had it running all right several times, so it's no a port issue. The polling may be the cause. I haven't experienced it, but the bahavior may also depend the exact state of the processes.

[optout21]

Added polling on bitcoind port

[optout21]

Changed to 'regtest' (instead of 'testnet4'), rebased to current master.

[vnprc]

Nice work @optout21! 💪