WebSockets in an IPv6 primary cluster fail

[MaxRink]

Describe the bug
We have Cluster that are either IPv6 only or IPv6 primary.
In those clusters the hub cant connect to the websocket of the workers.

2024-03-18T10:45:12Z INF server/controllers/socket.go:130 > Connecting to the worker at: url=ws://2a01:598:7e0:15::8:30001/ws?node=ship-lab-1-lx7qj&q=&worker=2a01%3A598%3A7e0%3A15%3A%3A8%3A30001
2024-03-18T10:45:12Z ERR server/controllers/socket.go:134 > WebSocket client dial: error="dial tcp: address 2a01:598:7e0:15::8:30001: too many colons in address"

Provide more information
On-Prem, upstream K8s 1.27.10
With Coil + Calico as CNI
To Reproduce
Steps to reproduce the behavior:

1. install kubeshark on an IPv6 primary cluster
2. Check the hub logs

Expected behavior
the hub can correctly use the websocket to the node agent

[mertyildiran]

@MaxRink thank you so much for reporting the issue!

We've realized that the method that builds the address of the workers does not have the IPv6 support. We patched it in a way that if it's IPv6, it will be like [2a01:598:7e0:15::8]:30001 instead of 2a01:598:7e0:15::8:30001. The fix will be there in the next release.