Forcefully delete pods when necessary (#4328)

Pods on down/unresponsive nodes can't be deleted without
--force --grace-period=0.

Fixes #4314

roles/kubernetes-apps/rotate_tokens/tasks/main.yml

@@ -44,5 +44,5 @@
   when: needs_rotation
 
 - name: Rotate Tokens | Delete pods in system namespace
-  command: "{{ bin_dir }}/kubectl --kubeconfig /etc/kubernetes/admin.conf delete pods -n kube-system --all"
+  command: "{{ bin_dir }}/kubectl --kubeconfig /etc/kubernetes/admin.conf delete pods -n kube-system --all --grace-period=0 --force"
   when: needs_rotation

roles/kubernetes/kubeadm/tasks/main.yml

@@ -133,7 +133,7 @@
     - kube-proxy
 
 - name: Restart all kube-proxy pods to ensure that they load the new configmap
-  shell: "{{ bin_dir }}/kubectl --kubeconfig {{ kube_config_dir }}/admin.conf delete pod -n kube-system -l k8s-app=kube-proxy"
+  shell: "{{ bin_dir }}/kubectl --kubeconfig {{ kube_config_dir }}/admin.conf delete pod -n kube-system -l k8s-app=kube-proxy --force --grace-period=0"
   delegate_to: "{{groups['kube-master']|first}}"
   run_once: true
   when: