v0.16.5

Changed

• Reject negative PyTorch replica and LWS worker indexes, and cap block-level segmentation at 10,000 subgroups to prevent unbounded PodGroup fan-out. #1816 davidLif

Fixed

• Fixed extended resources present on only a subset of nodes being reported as unavailable cluster-wide: ResourceVector.SetMax now grows the accumulator to the longer vector's length instead of silently dropping resource indices discovered after the first-iterated node, which caused pods requesting such resources to be rejected as unschedulable ("No node in the node-pool has X resources") depending on node map iteration order. #1851
• Scheduler snapshot now correctly captures the plugin configuration even when /get-snapshot is requested between scheduling cycles (previously the config field was written as null, causing snapshot-tool to panic on replay). #1885
• Fixed scheduler panic during reclaim when building eviction messages for jobs in root-level queues (ParentQueue empty) that reclaim across hierarchy branches. #1863
• Sanitize GPU-sharing volume names for pods with dots while preserving their ConfigMap references. #1728