upgrade k8s dependencies to support DRA

[movence]

Issue #, if available:
DRA is not handled correctly https://kubernetes.io/docs/concepts/scheduling-eviction/dynamic-resource-allocation/. Operator is not able to handle the new spec due to the old k8s deps.

Description of changes:
Upgrade k8s.io and sigs.k8s.io dependency versions which also bump go version to 1.24 due to one of upgraded dep.

Test
Deploying the operator with changes then use https://github.com/NVIDIA/k8s-dra-driver-gpu/tree/main/deployments/helm/nvidia-dra-driver-gpu

helm install dra ./nvidia-dra-driver-gpu --create-namespace --namespace nvidia-dra --set resources.gpus.enabled=false
NAME: dra
LAST DEPLOYED: Fri Jun 20 02:17:42 2025
NAMESPACE: nvidia-dra
STATUS: deployed
REVISION: 1
TEST SUITE: None
➜  helm git:(main) ✗ kubectl apply -f test.yaml                                                                                           
namespace/gpu-test1 created
resourceclaimtemplate.resource.k8s.io/single-gpu created
pod/pod1 created
pod/pod2 created

TODOs
- Need any additional changes required
- PR checks are all passing

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

[lisguo]

why was it necessary to dereference here?

[Paramadon]

LGTM!