uses model cpu offload instead of per-layer offload to speedup test ci

.github/workflows/pr_test_linux.yaml

@@ -93,7 +93,7 @@ jobs:
           source $(conda info --base)/etc/profile.d/conda.sh
           conda activate test_env || { echo "Failed to activate conda env"; exit 1; }
           which python
-          HF_TOKEN=${{ secrets.HF_TOKEN }} pytest -x tests/flux/test_flux_memory.py
+          HF_TOKEN=${{ secrets.HF_TOKEN }} pytest -s tests/flux/test_flux_memory.py
 
   test-flux-other:
     needs: build
@@ -108,7 +108,7 @@ jobs:
           source $(conda info --base)/etc/profile.d/conda.sh
           conda activate test_env || { echo "Failed to activate conda env"; exit 1; }
           which python
-          HF_TOKEN=${{ secrets.HF_TOKEN }} pytest -x tests/flux --ignore=tests/flux/test_flux_memory.py
+          HF_TOKEN=${{ secrets.HF_TOKEN }} pytest -s tests/flux --ignore=tests/flux/test_flux_memory.py
 
   test-sana:
     needs: build
@@ -123,7 +123,7 @@ jobs:
           source $(conda info --base)/etc/profile.d/conda.sh
           conda activate test_env || { echo "Failed to activate conda env"; exit 1; }
           which python
-          HF_TOKEN=${{ secrets.HF_TOKEN }} pytest -x tests/sana
+          HF_TOKEN=${{ secrets.HF_TOKEN }} pytest -s tests/sana
 
   clean-up:
     if: always() && (github.event_name != 'issue_comment' || needs.check-comment.outputs.should_run == 'true')