[Bug] Review reconnect logic for firehose

Bug report

When graph-node connects to firehose it may be provided a DNS name that resolves to multiple IPv4 and/or IPv6 addresses.

graph-node should explicitly try each of these available addresses. Of course if IPv6 is not configured in the networking stack, it would not try those. But if there are multiple IPv4 addresses, it should try each one in turn.

It appears, anecdotally, that graph-node does not specifically rotate between IPv4 addresses returned and gets stuck connecting to a single IPv4 address.

There are no logs for this, it is based on observed behaviour.

Relevant log output

No response

IPFS hash

No response

Subgraph name or link to explorer

No response

Some information to help us out

• Tick this box if this bug is caused by a regression found in the latest release.
• Tick this box if this bug is specific to the hosted service.
• I have searched the issue tracker to make sure this issue is not a duplicate.

OS information

None

Start by tracing graph-node's firehose connection and how it handles DNS names resolving to multiple IPv4 and IPv6 addresses. Reproduce or inspect behavior with a multi-address DNS result, including an environment without IPv6. Done means available addresses are tried in turn instead of remaining stuck on one IPv4 address.