CUDA Plugin Cleanup for Shared Kernel Helpers

[tianleiwu]

Description

This PR reduces the amount of CUDA plugin-specific compatibility code by moving reusable validation and attribute-reading logic into shared helper paths that work for both bundled and plugin builds. It also fills in a missing allocator hook in the EP adapter so plugin kernels can reuse the same initialization path as the in-tree CUDA EP, which simplifies maintenance and improves behavior parity. The follow-up changes update the CUDA plugin design doc to reflect the new shared-helper model and add focused plugin regression tests for the two runtime paths that changed most materially.

Summary of Changes

EP adapter and shared helper extraction

File	Change
ep/adapter/op_kernel_info.h	Adds OpKernelInfo::GetAllocator(OrtMemType) so adapter-based kernels can request device or CPU temp allocators in plugin builds.
cpu/tensor/scatter_nd.h	Extracts shape validation into scatter_nd_internal::ValidateShapes so the same logic can be reused outside the CPU ScatterND class.
cpu/tensor/space_depth_ops.h	Moves blocksize parsing, mode parsing, and dimension validation into space_depth_internal helpers that can be shared by CUDA kernels.

CUDA kernel cleanup and plugin parity

File	Change
cuda/tensor/scatter_nd.cc	Removes the plugin-only ScatterND validation duplicate and reuses the shared helper implementation.
cuda/tensor/scatter_nd.h	Drops the old conditional include split now that validation is shared through the common helper path.
cuda/tensor/space_depth_ops.h	Deletes the plugin-only SpaceToDepth/DepthToSpace reimplementation and inherits from the shared base/helper logic in all builds.
cuda/tensor/upsample.cc	Reuses the normal antialias lookup-table allocation/caching path in plugin builds via the new allocator adapter support.
cuda/tensor/upsample.h	Keeps the persistent device lookup-table member available in plugin builds as well.

Shared-provider and diagnostics alignment

File	Change
cpu/cpu_provider_shared.cc	Routes shared-provider ScatterND shape validation through the extracted helper.
provider_bridge_provider.cc	Updates the bridge-side ScatterND::ValidateShapes implementation to call the shared helper directly.
cuda/cudnn_common.h	Preserves the batch-norm epsilon warning path in plugin builds instead of suppressing it.
cuda/nn/conv.cc	Removes plugin-specific shortened cuDNN frontend errors so bundled and plugin builds both include frontend JSON in failures.
cuda/nn/conv_transpose.cc	Extends cuDNN frontend failures to include frontend JSON for easier debugging, matching the Conv behavior.

Documentation and regression coverage

File	Change
cuda_plugin_ep_design.md	Updates the design doc to reflect that ScatterND, SpaceDepth, and Upsample now use shared adapter-safe helper paths instead of plugin-only fallback branches.
test_cuda_plugin_ep.py	Adds plugin regression coverage for antialias Resize/Upsample and ScatterND, covering the new allocator-backed lookup-table path and the shared ScatterND validation helper.

Testing

• Build with onnxruntime_BUILD_CUDA_EP_AS_PLUGIN=ON and verify the affected CUDA provider sources compile without the removed plugin-only fallback paths.
• Run targeted CUDA provider coverage for ScatterND, SpaceToDepth/DepthToSpace, Resize/Upsample, Conv, and ConvTranspose in both plugin and bundled CUDA configurations.
• Confirm antialias upsample still initializes and uses the shared lookup table correctly in plugin builds.
• Run the new plugin tests for antialias Resize and ScatterND in onnxruntime/test/python/transformers/test_cuda_plugin_ep.py.
• Confirm cuDNN frontend failure paths now emit the same diagnostic detail in plugin and non-plugin builds.

Motivation and Context

The initial CUDA plugin enablement introduced several localized #ifdef BUILD_CUDA_EP_AS_PLUGIN branches and helper copies to get kernels compiling under the adapter path. This cleanup pays down that compatibility debt by extracting the truly shared pieces into reusable helpers and by teaching the adapter OpKernelInfo how to provide the allocators those kernels already expect. The result is less duplicated logic, fewer plugin-only code paths to keep in sync, and better debugging consistency between the plugin EP and the built-in CUDA EP.

Checklist

• Tests added/updated
• Documentation updated (if applicable)
• No breaking changes (or documented in description)

[Copilot]

Pull request overview

This PR reduces CUDA plugin-specific code paths by extracting shared validation/attribute-parsing helpers into adapter-safe headers, and by extending the EP adapter API to expose allocators needed by CUDA kernels—improving parity between bundled and plugin CUDA EP behavior.

Changes:

• Added ep::adapter::OpKernelInfo::GetAllocator(OrtMemType) to support allocator-dependent kernel initialization in plugin builds.
• Extracted shared helpers for ScatterND shape validation and SpaceToDepth/DepthToSpace attribute parsing/dimension validation, then reused them from CUDA/shared-provider code.
• Removed several plugin-only CUDA fallbacks and aligned cuDNN frontend diagnostics between plugin and bundled builds; updated docs and added plugin regression tests.

Reviewed changes

Copilot reviewed 15 out of 15 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
include/onnxruntime/ep/adapter/op_kernel_info.h	Exposes allocator access in adapter OpKernelInfo to enable shared CUDA initialization paths in plugin builds.
onnxruntime/core/providers/cpu/tensor/scatter_nd.h	Introduces scatter_nd_internal::ValidateShapes helper and routes ScatterND::ValidateShapes through it.
onnxruntime/core/providers/cuda/tensor/scatter_nd.cc	Switches CUDA ScatterND shape validation to shared helper and removes plugin-only duplicate validation.
onnxruntime/core/providers/cuda/tensor/scatter_nd.h	Removes conditional CPU include split now that validation is shared via helper.
onnxruntime/core/providers/cpu/cpu_provider_shared.cc	Routes shared-provider ScatterND validation through the extracted helper.
onnxruntime/core/providers/shared_library/provider_bridge_provider.cc	Updates bridge-side ScatterND validation to call the shared helper directly.
onnxruntime/core/providers/cpu/tensor/space_depth_ops.h	Extracts ReadBlocksize, ReadIsDCR, and dimension validation into space_depth_internal for reuse.
onnxruntime/core/providers/cuda/tensor/space_depth_ops.h	Removes plugin-only reimplementation and uses shared CPU base/helper logic for all builds.
onnxruntime/core/providers/cuda/tensor/upsample.h	Keeps persistent antialias lookup-table device buffer member available in plugin builds.
onnxruntime/core/providers/cuda/tensor/upsample.cc	Unifies plugin/bundled antialias lookup-table allocation/caching using allocator access.
onnxruntime/core/providers/cuda/cudnn_common.h	Preserves batch-norm epsilon warning behavior in plugin builds for parity.
onnxruntime/core/providers/cuda/nn/conv.cc	Aligns cuDNN frontend failure messages to always include frontend JSON (plugin parity).
onnxruntime/core/providers/cuda/nn/conv_transpose.cc	Extends cuDNN frontend failure diagnostics to include frontend JSON (matching Conv).
docs/cuda_plugin_ep/cuda_plugin_ep_design.md	Updates plugin design documentation to reflect the new shared-helper model and reduced #ifdef usage.
onnxruntime/test/python/transformers/test_cuda_plugin_ep.py	Adds targeted plugin regression tests for antialias Resize and ScatterND behavior.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 15 out of 15 changed files in this pull request and generated 2 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 15 out of 15 changed files in this pull request and generated 1 comment.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 15 out of 15 changed files in this pull request and generated 3 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[tianleiwu]

/azp run Linux QNN CI Pipeline

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).