Prerequisites

• I searched existing issues and found no duplicates
• I can reproduce this issue consistently
• This is not a security vulnerability (use Security Advisories instead)

Bug Description

DRA graduated to GA (resource.k8s.io/v1) in K8s 1.34, but the check only looks for v1beta1.

Impact

Blocking (cannot proceed)

Component

CLI (aicr)

Regression?

No, this is a new use case

Steps to Reproduce

1. Deploy a cluster running Kubernetes 1.34+ (DRA GA — resource.k8s.io/v1)
2. Install the NVIDIA DRA driver (nvidia-dra-driver-gpu v25.12.0)
3. Run validation: aicr validate --recipe recipe.yaml --snapshot snapshot.yaml --phase conformance
4. Observe dra-support check reports skipped:

SKIP reason="[INTERNAL] DRA API (resource.k8s.io/v1beta1) not available —
cluster may not support Dynamic Resource Allocation: skip"

Expected Behavior

The dra-support check should detect DRA via resource.k8s.io/v1 (GA) in addition to v1beta1 and report passed.

Actual Behavior

The check only queries resource.k8s.io/v1beta1, which was removed in K8s 1.34 when DRA graduated to GA. It skips, even though DRA is fully operational (confirmed by secure-accelerator-access check which creates ResourceClaims and runs GPU workloads via DRA successfully).

Environment

Kubernetes: v1.34.2 (OKE)
NVIDIA DRA Driver: v25.12.0
GPU: NVIDIA GB200
AICR: dev

Command / Request Used

No response

Logs / Error Output

Additional Context

No response