Make connecting to evm code more robust

[ryardley]

          _:hammer_and_wrench: Refactor suggestion_

Handle potential partial failures when attaching to multiple chains

In the loop attaching EnclaveSolReader and CiphernodeRegistrySol to each enabled chain, any failure in one chain will cause the entire setup to fail due to the use of await?. Consider handling errors for each chain individually, allowing the cipher node to continue operating with the chains that successfully attached.

You might refactor the loop to collect and log errors without stopping the entire setup:

for chain in config
    .chains()
    .iter()
    .filter(|chain| chain.enabled.unwrap_or(true))
{
    let rpc_url = &chain.rpc_url;

    if let Err(e) = EnclaveSolReader::attach(&bus, rpc_url, &chain.contracts.enclave).await {
        // Log the error without returning
        eprintln!("Failed to attach EnclaveSolReader for chain {}: {:?}", chain.name, e);
        continue;
    }

    if let Err(e) = CiphernodeRegistrySol::attach(&bus, rpc_url, &chain.contracts.ciphernode_registry).await {
        // Log the error without returning
        eprintln!("Failed to attach CiphernodeRegistrySol for chain {}: {:?}", chain.name, e);
        continue;
    }
}

This approach ensures that issues with a single chain do not prevent the cipher node from initializing with other available chains.

Originally posted by @coderabbitai[bot] in #156 (comment)