NVIDIA GPU operator validator pod fails on OpenShift 4 using crun

Issue

• When using the NVIDIA GPU Operator v25.3.2 on Red Hat OpenShift Container Platform 4.18.22 -or- 4.18.23, the nvidia-operator-validator fails with the following initialization error message:

  The nvidia-operator-validator pod in Init:CreateContainerError - error executing hook `/usr/local/nvidia/toolkit/nvidia-container-runtime-hook` (exit code: 1)  
  

  This issue will not be present on clusters before OpenShift Container Platform 4.18.22 or on clusters running above 4.18.23. They are only these two versions that are impacted.

• The impact is only related to clusters where the container runtime is crun. Clusters that use runc are not impacted.