[cifmw_helpers] Fix CRC certificate renewal wait using kubeconfig

The oc login command in crc_start.yml was missing the -p (password)
flag, causing every retry to fail with 401 Unauthorized regardless
of cluster health. This code path was never exercised until ~2026-06-03
when the CRC image certificates expired, breaking all CRC-based
molecule jobs.

Replace oc login with oc get nodes using the CRC kubeconfig file.
This avoids the OAuth dependency during certificate renewal and does
not require credentials. Also increase retries from 90 to 150 (~25
minutes) to cover the observed recovery time.

Signed-off-by: Vito Castellano <vcastell@redhat.com>

Author: Valkyrie00

roles/cifmw_helpers/tasks/crc_start.yml

@@ -6,15 +6,13 @@
   register: _crc_output
   ignore_errors: true  # noqa: ignore-errors
 
-- name: Login to the OpenShift when certificate is expired
+- name: Wait for cluster to stabilize after certificate renewal
   when: "'Kubelet serving certificate has expired' in _crc_output.stderr"
   ansible.builtin.command: >
-    oc login
-    -u kubeadmin
-    https://api.crc.testing:6443
-    --insecure-skip-tls-verify
-  register: _openshift_login
-  until: _openshift_login.rc == 0
-  retries: 90
+    oc get nodes
+    --kubeconfig={{ ansible_user_dir }}/.crc/machines/crc/kubeconfig
+  register: _cluster_check
+  until: _cluster_check.rc == 0
+  retries: 150
   delay: 10
   changed_when: false