Refactor cuda packaging pipeline (microsoft#22542)

### Description
<!-- Describe your changes. -->



### Motivation and Context
<!-- - Why is this change required? What problem does it solve?
- If it fixes an open issue, please link to the issue here. -->

Author: jchen351

tools/ci_build/github/azure-pipelines/build-perf-test-binaries-pipeline.yml

@@ -34,11 +34,8 @@ stages:
 # build Python packages
 # Linux GPU only
 - ${{ if parameters.BuildPythonPackages }}:
-  - template: templates/py-packaging-stage.yml
+  - template: stages/py-gpu-packaging-stage.yml
     parameters:
       enable_linux_gpu: true
-      enable_linux_cpu: false
-      enable_windows_cpu: false
       enable_windows_gpu: false
-      enable_mac_cpu: false
-      enable_linux_arm: false
+      cuda_version: 12.2

tools/ci_build/github/azure-pipelines/py-cuda-alt-packaging-pipeline.yml

@@ -0,0 +1,27 @@
+trigger: none
+
+parameters:
+  - name: enable_linux_cuda
+    type: boolean
+    default: true
+
+  - name: enable_windows_cuda
+    type: boolean
+    default: true
+
+  - name: cmake_build_type
+    type: string
+    default: 'Release'
+    values:
+      - Debug
+      - Release
+      - RelWithDebInfo
+      - MinSizeRel
+
+stages:
+  - template: stages/py-gpu-packaging-stage.yml
+    parameters:
+      enable_linux_cuda: ${{ parameters.enable_linux_cuda }}
+      enable_windows_cuda: ${{ parameters.enable_windows_cuda }}
+      cmake_build_type: ${{ parameters.cmake_build_type }}
+      cuda_version: '11.8'

tools/ci_build/github/azure-pipelines/py-cuda-packaging-pipeline.yml

@@ -1,12 +1,20 @@
 trigger: none
-
+# The `resources` specify the location and version of the 1ES PT.
+resources:
+  repositories:
+  - repository: 1esPipelines
+    type: git
+    name: 1ESPipelineTemplates/1ESPipelineTemplates
+    ref: refs/tags/release
 parameters:
-  - name: enable_linux_gpu
+  - name: enable_linux_cuda
     type: boolean
     default: true
-  - name: enable_windows_gpu
+
+  - name: enable_windows_cuda
     type: boolean
     default: true
+
   - name: cmake_build_type
     type: string
     default: 'Release'
@@ -15,28 +23,22 @@ parameters:
       - Release
       - RelWithDebInfo
       - MinSizeRel
-  - name: cuda_version
-    type: string
-    default: '12.2'
-    values:
-      - 11.8
-      - 12.2
-  - name: SpecificArtifact
-    displayName: Use Specific Artifact
-    type: boolean
-    default: false
 
-  - name: BuildId
-    displayName: Specific Artifact's BuildId
-    type: string
-    default: '0'
+extends:
+  # The pipeline extends the 1ES PT which will inject different SDL and compliance tasks.
+  # For non-production pipelines, use "Unofficial" as defined below.
+  # For productions pipelines, use "Official".
+  template: v1/1ES.Official.PipelineTemplate.yml@1esPipelines
+  parameters:
+    # Update the pool with your team's 1ES hosted pool.
+    pool:
+      name: 'onnxruntime-Win-CPU-2022'  # Name of your hosted pool
+      os: windows  # OS of the image. This value cannot be a variable. Allowed values: windows, linux, macOS
 
-stages:
-  - template: stages/py-cuda-packaging-stage.yml
-    parameters:
-      enable_linux_gpu: ${{ parameters.enable_linux_gpu }}
-      enable_windows_gpu: ${{ parameters.enable_windows_gpu }}
-      cmake_build_type: ${{ parameters.cmake_build_type }}
-      cuda_version: ${{ parameters.cuda_version }}
-      SpecificArtifact: ${{ parameters.SpecificArtifact }}
-      BuildId: ${{ parameters.BuildId }}
+    stages:
+      - template: stages/py-gpu-packaging-stage.yml
+        parameters:
+          enable_linux_cuda: ${{ parameters.enable_linux_cuda }}
+          enable_windows_cuda: ${{ parameters.enable_windows_cuda }}
+          cmake_build_type: ${{ parameters.cmake_build_type }}
+          cuda_version: '12.2'

tools/ci_build/github/azure-pipelines/py-dml-packaging-pipeline.yml

@@ -0,0 +1,18 @@
+trigger: none
+
+parameters:
+  - name: cmake_build_type
+    type: string
+    default: 'Release'
+    values:
+      - Debug
+      - Release
+      - RelWithDebInfo
+      - MinSizeRel
+
+stages:
+  - template: stages/py-gpu-packaging-stage.yml
+    parameters:
+      enable_windows_dml: true
+      cmake_build_type: ${{ parameters.cmake_build_type }}
+      publish_symbols: true

tools/ci_build/github/azure-pipelines/py-packaging-pipeline.yml

@@ -4,21 +4,11 @@ parameters:
   type: boolean
   default: true
 
-- name: enable_linux_gpu
-  displayName: 'Whether Linux GPU package is built.'
-  type: boolean
-  default: true
-
 - name: enable_windows_cpu
   displayName: 'Whether Windows CPU package is built.'
   type: boolean
   default: true
 
-- name: enable_windows_gpu
-  displayName: 'Whether Windows GPU package is built.'
-  type: boolean
-  default: true
-
 - name: enable_mac_cpu
   displayName: 'Whether Mac CPU package is built.'
   type: boolean
@@ -74,12 +64,10 @@ parameters:
 trigger: none
 
 stages:
-- template: templates/py-packaging-stage.yml
+- template: stages/py-cpu-packaging-stage.yml
   parameters:
-    enable_linux_gpu: ${{ parameters.enable_linux_gpu }}
     enable_linux_cpu: ${{ parameters.enable_linux_cpu }}
     enable_windows_cpu: ${{ parameters.enable_windows_cpu }}
-    enable_windows_gpu: ${{ parameters.enable_windows_gpu }}
     enable_mac_cpu: ${{ parameters.enable_mac_cpu }}
     enable_linux_arm: ${{ parameters.enable_linux_arm }}
     enable_windows_arm64_qnn: ${{ parameters.enable_windows_arm64_qnn }}
@@ -90,3 +78,4 @@ stages:
     cmake_build_type: ${{ parameters.cmake_build_type }}
     qnn_sdk_version: ${{ parameters.qnn_sdk_version }}
     publish_symbols: true
+

tools/ci_build/github/azure-pipelines/stages/jobs/py-linux-cuda-package-test-job.yml

@@ -57,15 +57,15 @@ jobs:
       - checkout: self
       - task: DownloadPipelineArtifact@2
         inputs:
-          artifact: 'drop-linux-gpu-x86_64'
-          targetPath: '$(Build.SourcesDirectory)/drop-linux-gpu-x86_64'
+          artifact: 'linux_gpu_wheel_x86_64'
+          targetPath: '$(Build.SourcesDirectory)/linux_gpu_wheel_x86_64'
           ${{ if ne(parameters.build_id, 'latest') }}:
             buildType: 'specific'
             project: '${{ parameters.project }}'
             pipeline: '${{ parameters.pipeline }}'
             buildVersionToDownload: 'specific'
             buildId: '${{ parameters.build_id }}'
-        displayName: 'Download Build Artifacts - drop-linux-gpu-x86_64'
+        displayName: 'Download Build Artifacts - linux_gpu_wheel_x86_64'
 
       - task: DownloadPipelineArtifact@2
         inputs:
@@ -82,7 +82,7 @@ jobs:
       - bash: |
           set -e -x
           ls $(Build.SourcesDirectory)
-          mv "$(Build.SourcesDirectory)/drop-linux-gpu-x86_64" $(Build.BinariesDirectory)/${{parameters.cmake_build_type}}
+          mv "$(Build.SourcesDirectory)/linux_gpu_wheel_x86_64" $(Build.BinariesDirectory)/${{parameters.cmake_build_type}}
           mv "$(Build.SourcesDirectory)/onnxruntime_gpu" "$(Build.BinariesDirectory)/whl"
           cp -r "$(Build.BinariesDirectory)/whl" $(Build.BinariesDirectory)/tmp
           find "$(Build.BinariesDirectory)/tmp" -name '*.whl' -exec bash -c 'unzip -d "${1%.*}" "$1"' _ {} \;