Cannot resume_pod with gpu_count=0

[vanduc2514]

Describe the bug
I use the following code to create a pod with 1 GPU initially, and then resume it with 0 GPU (because creating a pod with 0 gpu is not possible). The pod resumes but with 1

pod_id = runpod.create_pod(
    name="pod",
    image_name=runpod_image_name,
    cloud_type="SECURE",
    network_volume_id=network_volume_id,
    volume_mount_path="/volume-storage",
    docker_args="tail -f /dev/null",
    gpu_count=1,
    gpu_type_id="NVIDIA RTX A4000",
    container_disk_in_gb=50
)["id"]

runpod.stop_pod(pod_id)

runpod.resume_pod(
    pod_id=pod_id,
    gpu_count=0 # FIXME: this does not work
)

To Reproduce

1. Create a python script and import runpod
2. Set api_key
3. Create a pod with minium 1 GPU
4. Stop the created pod
5. Resume the stopped pod with 0 GPU

Expected behavior
Pod resume with 0 GPU

Screenshots

Desktop (please complete the following information):

• OS: Linux
• Browser edge
• Version 22

Additional context

If I stop the pod and go to Web UI, then manually start the pod with 0 GPU then the pod can be resumed with 0 GPU

pod_id = "value from Web UI"

runpod.stop_pod(pod_id)

runpod.resume_pod(
    pod_id=pod_id,
    gpu_count=0 
)