feat: Log bind address at startup

[stbrody]

With this change at startup there's a message like

  2024-07-24T21:40:00.705970Z  INFO ceramic_one: starting api server at address 127.0.0.1:5101
    at one/src/lib.rs:575

It does make me wonder though if we should separate bind_address into separate args for the port to use and the network interface to bind to. It's kind of weird to print 127.0.0.1:5101 when presumably we also accept connections that don't just come from localhost. Really it's the port that's going to be the most interesting part for most people

[linear]

AES-252 ceramic-one should log what API port it's using at startup

[dav1do]

For some reason I thought we already did this but it's only the p2p address. I think accepting the socket address as one parameter is fine personally, but it might be nice to actually make the CLI flag a SocketAddr instead of a string. It doesn't really change much, other than getting an error earlier e.g.

❯ ./target/debug/ceramic-one daemon --network local --local-network-id 0 --bind-address 127.0.0:5101
  2024-07-25T17:43:18.790913Z  INFO ceramic_one: service__name: "ceramic-one", version: "0.29.0", build: "git:570d299f", instance_id: "beautiful-able", exe_hash: "uAAUAdebugg"
    at one/src/lib.rs:343

...

  2024-07-25T17:43:18.974446Z  INFO ceramic_api::server: Shutting down insert task.
    at api/src/server.rs:347

  2024-07-25T17:43:18.974709Z ERROR ceramic_one: Error running command: invalid socket address syntax
    at one/src/main.rs:15

Error: invalid socket address syntax

vs

❯ ./target/debug/ceramic-one daemon --network local --local-network-id 0 --bind-address 127.0.0:5101
error: invalid value '127.0.0:5101' for '--bind-address <BIND_ADDRESS>': invalid socket address syntax

[stbrody]

yeah that does look like a nice improvement