Kubernetes v1.37 adds alpha scheduler preemption for in-place pod resize

Kubernetes v1.37 introduced scheduler preemption for in-place pod resize as an alpha feature behind the InPlacePodVerticalScalingSchedulerPreemption gate. Previously, a running pod whose scale-up exceeded node headroom was marked Deferred and could stay blocked indefinitely. The new mechanism lets the scheduler evict lower-priority pods on the same node to free capacity, then the kubelet actuates the resize without restarting the pod. Preemption is scoped to the pod's assigned node, and administrators can disable resize preemption per node via the podPreemptionPolicy field.