Set custom xpk imagePullSecret (#1559)

Cherry-pick of changes (39baf57, 293f0dc, 4b233e6) to xpk composite
action to customize reference to imagePullSecret in xpk workload

(n.b. this seems to be necessary in `main` to enable the ngc release
branch to use it in a workflow dispatch event)

Author: aybchan

.github/actions/gke-xpk/action.yml

@@ -47,6 +47,11 @@ inputs:
     required: false
     default: ghcr.io/nvidia/jax:latest
     type: string
+  IMAGE_PULL_SECRET_NAME:
+    description: 'Name of k8s Secret resource for registry ImagePullSecret'
+    required: false
+    default: jax-toolbox-ghcr
+    type: string
   COMMAND:
     description: 'Command to run in main container on JobSet start up'
     required: false
@@ -113,9 +118,10 @@ runs:
   - name: Apply XPK workload create patch
     shell: bash -x -e -u {0}
     run: |
-      git apply --unsafe-paths .github/gke-workflow/xpk/tcpxo_decorator.patch --directory ${WORKLOAD_NAME}/xpk
-      git apply --unsafe-paths .github/gke-workflow/xpk/docker_resources.patch --directory ${WORKLOAD_NAME}/xpk
-      git apply --unsafe-paths .github/gke-workflow/xpk/workload.patch --directory ${WORKLOAD_NAME}/xpk
+      sed -i 's/{{ IMAGE_PULL_SECRET_NAME }}/${{ inputs.IMAGE_PULL_SECRET_NAME }}/g' .github/gke-workflow/xpk/${{ inputs.XPK_VERSION}}/workload.patch 
+      git apply --unsafe-paths .github/gke-workflow/xpk/${{ inputs.XPK_VERSION}}/tcpxo_decorator.patch --directory ${WORKLOAD_NAME}/xpk
+      git apply --unsafe-paths .github/gke-workflow/xpk/${{ inputs.XPK_VERSION}}/docker_resources.patch --directory ${WORKLOAD_NAME}/xpk
+      git apply --unsafe-paths .github/gke-workflow/xpk/${{ inputs.XPK_VERSION}}/workload.patch --directory ${WORKLOAD_NAME}/xpk
   
   - name: Set workload commands
     shell: bash -x -e -u {0}

.github/gke-workflow/xpk/blueprint.patch renamed to .github/gke-workflow/xpk/v0.8.0/blueprint.patch

@@ -7,7 +7,7 @@ index a466a5c..8a5b99e 100644
                - operator: "Exists"
                  key: nvidia.com/gpu
 +              imagePullSecrets:
-+              - name: jax-toolbox-ghcr
++              - name: {{ IMAGE_PULL_SECRET_NAME }}
                containers:
                {container}
  """

.github/gke-workflow/xpk/docker_resources.patch renamed to .github/gke-workflow/xpk/v0.8.0/docker_resources.patch

@@ -24,10 +24,20 @@ jobs:
       NUM_NODES: 2
 
     steps:
+    - name: Login to GitHub Container Registry
+      uses: docker/login-action@v3
+      with:
+        registry: ghcr.io
+        username: ${{ github.repository_owner }}
+        password: ${{ secrets.GITHUB_TOKEN }}
+    - name: K8s GHCR store and delete token
+      id: store-token
+      uses: ./.github/actions/store-delete-k8s-ghcr
     - name: Run XPK workload on cluster
       uses: ./.github/actions/gke-xpk
       with:
         IMAGE: ${{ env.MAXTEXT_IMAGE }}
+        IMAGE_PULL_SECRET_NAME: ${{ steps.store-token.outputs.token-name }}
         WORKLOAD_NAME_PREFIX: ${{ env.WORKLOAD_NAME_PREFIX }}
         COMMAND: |
           export NCCL_NET_PLUGIN=/opt/hpcx/nccl_rdma_sharp_plugin/lib/libnccl-net.so;