Netpol allow-kube-apiserver-egress-alloy causes failure to start if kubernetes.default.svc Endpoints change after chart rendering
The [current implementation of the `allow-kube-apiserver-egress` Network Policy](https://repo1.dso.mil/big-bang/product/packages/alloy/-/blob/main/chart/templates/bigbang/networkpolicies/allow-kube-apiserver-egress.yaml?ref_type=heads#L14) performs a lookup of the Endpoints for the kubernetes.default.svc.cluster.local Service, allowing the Alloy pods to talk to the kube-api via those Endpoints. However, the lookup is only valid at the time of Helm template rendering; if the Endpoints are updated after the NetPol is rendered, this can result in a state where the Alloy pods are not able to start correctly due to not being able to reach the kube-api successfully.
issue