onnxruntime

a8424e67 - Update CUDA TopK kernel registration to opset 24 with BFloat16 support (#27735)

Update CUDA TopK kernel registration to opset 24 with BFloat16 support (#27735) - [x] Cap existing CUDA TopK kernel to versioned [11, 23] and add opset 24 - [x] Add BFloat16 support for CUDA TopK opset 24 (topk_impl_bf16.cu, helpers, NumericLimits) - [x] Add BFloat16 test cases for TopK opset 24 - [x] Fix CUB build error: map BFloat16 → __nv_bfloat16 for BlockRadixSort and DeviceRadixSort - [x] Add CubSortType trait in topk_impl.cuh - [x] Update RadixTopK kernel to use CubSortType for BlockRadixSort - [x] Update TopKImpl DeviceRadixSort calls to use CubSortType pointers <!-- START COPILOT CODING AGENT TIPS --> --- 💬 Send tasks to Copilot coding agent from [Slack](https://gh.io/cca-slack-docs) and [Teams](https://gh.io/cca-teams-docs) to turn conversations into code. Copilot posts an update in your thread when it's finished. --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: tianleiwu <30328909+tianleiwu@users.noreply.github.com>