Add env variable to signal skip vfio-pci unbind

[karthikvetrivel]

Relevant PR: NVIDIA/k8s-driver-manager#146

Description

Pass GPU_WORKLOAD_CONFIG environment variable to k8s-driver-manager init container in vfio-manager DaemonSet to prevent unnecessary GPU unbind/rebind operations during rolling updates.

Problem

During rolling updates of the vfio-manager DaemonSet, k8s-driver-manager unconditionally unbinds all GPUs from vfio-pci on startup. When the desired state is already vfio-pci binding, this causes unnecessary disruption to active VM workloads using GPU passthrough (KubeVirt, Kata Containers).

Design Rationale

We know that vfio-manager only runs on vm-passthrough nodes: The DaemonSet's nodeSelector requires nvidia.com/gpu.deploy.vfio-manager: "true", which is only set for gpuWorkloadConfigVMPassthrough nodes. This is true regardless of whether the workload config comes from an explicit node label or sandboxWorkloads.defaultWorkload.

Checklist

• No secrets, sensitive information, or unrelated changes
• Lint checks passing (make lint)
• Generated assets in-sync (make validate-generated-assets)
• Go mod artifacts in-sync (make validate-modules)
• Test cases are added for new code paths