Proxy panics on startup when `DOWNSTREAM_HASHRATE` env var is set

[Anshumancanrock]

Description

Setting the DOWNSTREAM_HASHRATE environment variable causes the proxy to panic immediately on startup with:

Root Cause

I think there is a copy-paste error in src/config.rs where the listening_addr field incorrectly reads from DOWNSTREAM_HASHRATE instead of a dedicated env var.

let listening_addr = args.listening_addr.or(config.listening_addr).or_else(|| {
    std::env::var("DOWNSTREAM_HASHRATE")  // should be a dedicated env var, e.g. LISTENING_ADDR
        .ok()
        .and_then(|s| s.parse().ok())
});

Since the return type is Option<String>, parsing a hashrate string like "100T" always succeeds. The value is then passed to TcpListener::bind, which fails to parse it as a SocketAddr and panics.

This also means there is currently no working environment variable to configure the downstream listening address.

Steps to Reproduce

DOWNSTREAM_HASHRATE=100T ./dmnd-client

Expected Behaviour

Setting DOWNSTREAM_HASHRATE should only affect the hashrate configuration and have no impact on the listening address.

Proposed Fix

Replace "DOWNSTREAM_HASHRATE" with an appropriate env var name (e.g. "LISTENING_ADDR") in the listening_addr resolution block.