[release-1.14] Fix hami vGPU scheduling failure in large and medium-scale clusters

[Copilot]

Cherry-pick of aec04d7 from master to release-1.14. The handshake annotation mechanism (volcano.sh/node-vgpu-handshake) caused vGPU scheduling failures in medium/large clusters due to timing inconsistencies between the scheduler and hami-dp.

Changes

• pkg/scheduler/api/devices/config/vgpu.go: Remove VolcanoVGPUHandshake constant
• pkg/scheduler/api/devices/nvidia/vgpu/device_info.go: Replace handshake annotation check in NewGPUDevices with direct node.Status.Allocatable checks for vgpu-number, vgpu-cores, and vgpu-memory
• pkg/scheduler/api/devices/nvidia/vgpu/utils.go: Remove patchNodeAnnotations helper (no longer referenced)

Before → After

// Before: gated on a handshake annotation that could get stuck
handshake, ok := node.Annotations[deviceconfig.VolcanoVGPUHandshake]
if !ok {
    return nil
}
// ... timing-based state machine using "Requesting_" / "Deleted_" strings

// After: check allocatable resources directly
gpuNumberRes, gpuNumberExists := node.Status.Allocatable[v1.ResourceName(deviceconfig.VolcanoVGPUNumber)]
if !gpuNumberExists || gpuNumberRes.Value() == 0 {
    return nil
}
// ... similar checks for vgpu-cores and vgpu-memory

[volcano-sh-bot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please ask for approval from jessestutler. For more information see the Kubernetes Code Review Process.

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

• pkg/scheduler/OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment