e2e: alpine image pull failure on CRI-O 1.34+ due to short name enforcement

[RamLavi]

What happend:
The install-cni-plugins daemonset in the e2e test setup fails to start because CRI-O 1.34 refuses to pull the alpine image due to short name mode enforcement.

$ kubectl get pod -n kube-system -l name=cni-plugins
NAME                          READY   STATUS             RESTARTS   AGE
install-cni-plugins-2rpm4     0/1     ImagePullBackOff   0          9m
install-cni-plugins-4njl9     0/1     ImagePullBackOff   0          9m

$ kubectl describe pod -n kube-system install-cni-plugins-2rpm4
Events:
  Type     Reason     Age   From     Message
  ----     ------     ----  ----     -------
  Warning  Failed     6m    kubelet  Failed to pull image "alpine": short name mode is enforcing, but image name alpine:latest returns ambiguous list
  Warning  Failed     6m    kubelet  Error: ErrImagePull

The issue is in e2e/templates/cni-install.yml.j2 which uses image: alpine (unqualified short name).

What you expected to happen:
The install-cni-plugins pods should start successfully...

How to reproduce it (as minimally and precisely as possible):
Spin up a Kubernetes 1.34 cluster with CRI-O 1.34 (e.g., using kubevirtci k8s-1.34 provider)
Deploy the multus e2e test environment using e2e/templates/cni-install.yml.j2
Observe the pods fail with ImagePullBackOff
example

Anything else we need to know?:
CRI-O 1.34 enforces "short name mode" by default, which refuses to pull images with unqualified names like alpine because they're ambiguous (could resolve to docker.io, quay.io, etc.).

temporary fix:

kubectl set image daemonset/install-cni-plugins -n kube-system install-cni-plugins=docker.io/library/alpine

Environment:

• Multus version: v4.2.3 (commit 2ecd4f4) (but also on master)
  image path and image ID (from 'docker images')
• Kubernetes version (use kubectl version):

  Client Version: v1.34.1
  Server Version: v1.34.1

• Primary CNI for Kubernetes cluster: Calico v3.26.5
• OS (e.g. from /etc/os-release): CentOS Stream 9 (kubevirtci node image)
• Container Runtime: CRI-O 1.34
• File of '/etc/cni/net.d/' N/A (issue occurs before CNI setup completes)
• File of '/etc/cni/multus/net.d' N/A
• NetworkAttachment info (use kubectl get net-attach-def -o yaml) N/A
• Target pod yaml info (with annotation, use kubectl get pod <podname> -o yaml) The failing pod is from e2e/templates/cni-install.yml.j2 daemonset
• Other log outputs (if you use multus logging) N/A (image pull failure, container never starts)