Add SUBDIR mapping for vgpu rhel9.x targets

[mvalsecchi-nv]

Close #494 .

On main, rhel9.6 targets resolve to the wrong subdirectory (rhel9.6/ instead of rhel9/), as reported in #486.

Steps from cloud-native docs.

Before the fix (on main):

 $ git log -1 --pretty=oneline
b36127925fece95b9e2edcc55d0dcd8e42636cdb (HEAD -> main, origin/main, origin/HEAD) Merge pull request #564 from NVIDIA/dependabot/github_actions/renovatebot/github-action-44.2.5
 $ export PRIVATE_REGISTRY=my.private.registry.example.com VGPU_HOST_DRIVER_VERSION=570.211.01 OS_TAG=rhel9.6
 $ GPU_HOST_DRIVER_VERSION=${VGPU_HOST_DRIVER_VERSION} IMAGE_NAME=${PRIVATE_REGISTRY}/vgpu-manager make build-vgpuhost-${OS_TAG}
DOCKER_BUILDKIT=1 \
        docker  build --pull \
                        --output=type=image,push=false \
                        --platform=linux/amd64 \
                        --tag my.private.registry.example.com/vgpu-manager:570.211.01-rhel9.6 \
                        --build-arg DRIVER_BRANCH="570" \
                        --build-arg DRIVER_VERSION="570.211.01" \
                        --build-arg GOLANG_VERSION="1.25.6" \
                        --build-arg CVE_UPDATES="" \
                        --build-arg CUDA_VERSION="" \
                         \
                        --file /home/user/gpu-driver-container/vgpu-manager/rhel9.6/Dockerfile \
                        /home/user/gpu-driver-container/vgpu-manager/rhel9.6
[+] Building 0.0s (0/0)                                                                                                                                                                        docker:default
ERROR: failed to build: unable to prepare context: path "/home/user/gpu-driver-container/vgpu-manager/rhel9.6" not found
make: *** [Makefile:290: build-vgpuhost-rhel9.6] Error 1

After the fix:

$ export PRIVATE_REGISTRY=my.private.registry.example.com VGPU_HOST_DRIVER_VERSION=570.211.01 OS_TAG=rhel9.6
 $ VGPU_HOST_DRIVER_VERSION=${VGPU_HOST_DRIVER_VERSION} IMAGE_NAME=${PRIVATE_REGISTRY}/vgpu-manager make build-vgpuhost-${OS_TAG}
[+] Building 0.7s (15/15) FINISHED
 => => writing image sha256:25bd786d611d404401e1deb2c10b4df172d8459b2dc5e7d50e2aad2419f0f9bf
 => => naming to my.private.registry.example.com/vgpu-manager:570.211.01-rhel9.6

$ docker image ls --no-trunc --format '{{.ID}}' my.private.registry.example.com/vgpu-manager:570.211.01-rhel9.6
sha256:25bd786d611d404401e1deb2c10b4df172d8459b2dc5e7d50e2aad2419f0f9bf

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.