fix: restrict network policies closer to what is minimum required access

[JoeHCQ1]

This restricts the network policies. It does not close #7 as the following changes are still required:

1. Once clustering is enabled Enable Confluence Clustering by default #26, the network policy enabling clustering must be confirmed. If multicast is required to create the cluster (alternative is we hard-code to pod IPs) then the policies as-is are likely to require changes.
2. As monitoring is shown to be working properly Must implement monitors for each application metrics endpoint using it's built-in chart monitors, the Package CR monitor key, or manual monitors in the config chart. #12, the policies related to Prometheus will need validated.
3. Once Synchrony is enabled, like node clustering, the policies which enable these comms will need added and validated.
4. Once restrictive egress is enabled in UDS core Lock down "egress anywhere" policies for known external services uds-core#558 the egress to postgresql needs restricted.

That said, for the capabilities which were known to exist before this work started, the network policies have been restricted.

[JoeHCQ1]

Closing this for now. #34 made it largely irrelevant. I can still restrict the intra namespace policy but I don't want to do that until I have clustering working so I can confirm that my policies are not breaking anything.