Add automatic worker discovery to console

[EdsonPetry]

The console now always discovers workers via GetClusterWorkers RPC. This change enables the console to automatically discover workers over the network. Instrumentation now works like so:

// console/examples/console_worker.rs
let localhost_resolver = Arc::new(LocalhostWorkerResolver { ports: ..., });
let worker = Worker::default();

Server::builder()
    .add_service(worker.with_observability_service(resolver)) // users now pass in their WorkerResolver impl
    .add_service(worker.into_flight_server())
    .serve(SocketAddr::new(
        IpAddr::V4(Ipv4Addr::LOCALHOST),
        args.port.unwrap_or(DEFAULT_WORKER_PORT),
    )).await?;

Summary

• Add a new GetClusterWorkers gRPC to the observability service that returns all known worker URLs by delegating to the existing WorkerResolver trait
• Replace the console's manual --cluster-ports CLI flag with a --connect seed URL (defaults to port 9001)
• Wire up the observability service in the benchmarks EC2 worker and all console examples with a user-provided WorkerResolver
• Export DEFAULT_WORKER_PORT (9001) as the canonical default across the codebase
• Add a console/README.md and update all example docs for the new auto-discovery workflow

[gabotechs]

👍 nice! no comments, this looks good