Workaround problems in Cilium v1.14 partial kube-proxy replacement

* With Cilium v1.14, Cilium's kube-proxy partial mode changed to
either be enabled or disabled (not partial). This somtimes leaves
Cilium (and the host) unable to reach the kube-apiserver via the
in-cluster Kubernetes Service IP, until the host is rebooted
* As a workaround, configure Cilium to rely on external DNS resolvers
to find the IP address of the apiserver. This is less portable
and less "clean" than using in-cluster discovery, but also what
Cilium wants users to do. Revert this when the upstream issue
https://github.com/cilium/cilium/issues/27982 is resolved

google-cloud/flatcar-linux/kubernetes/bootstrap.tf

@@ -1,6 +1,6 @@
 # Kubernetes assets (kubeconfig, manifests)
 module "bootstrap" {
-  source = "git::https://github.com/poseidon/terraform-render-bootstrap.git?ref=ae571974b0b7a7fcd93c12f635fb8f2d6808ac51"
+  source = "git::https://github.com/poseidon/terraform-render-bootstrap.git?ref=720adbeb43a8b2860bf92544600f1fd4f0d2a907"
 
   cluster_name          = var.cluster_name
   api_servers           = [format("%s.%s", var.cluster_name, var.dns_zone)]