Add retries to initial image fetch

Since we're fetching containers from a remote location, if there is any
sort of connectivity issues while deploying, we end up failing the
deploy. We should retries these calls to improve deployment reliability.

Change-Id: Ia67f46f418b05bdf3ad797aab496c9b430c47650
Closes-Bug: #1878615
(cherry picked from commit c071d14d32)

deployment/deprecated/docker/docker-baremetal-ansible.yaml

@@ -192,6 +192,8 @@ outputs:
             - name: Pre-fetch all the containers
               become: true
               shell: "docker pull {{ container_image }}"
+              retries: 5
+              delay: 5
               loop_control:
                 loop_var: container_image
               loop: "{{ lookup('file', tripleo_role_name + '/docker_config.yaml', errors='ignore') | default('{}', True) | from_yaml | recursive_get_key_from_dict(key='image') | unique }}"

deployment/podman/podman-baremetal-ansible.yaml

@@ -149,6 +149,8 @@ outputs:
               podman_image:
                   name: "{{ container_image }}"
                   validate_certs: false
+              retries: 5
+              delay: 5
               loop_control:
                   loop_var: container_image
               loop: "{{ lookup('file', tripleo_role_name + '/docker_config.yaml', errors='ignore') | default('{}', True) | from_yaml | recursive_get_key_from_dict(key='image') | unique }}"