[CI][AMD][BugFix] Add torch.cuda.set_device to test_punica_ops so punica kernels execute on same device as tensor (vllm-project#34985)

Signed-off-by: Randall Smith <Randall.Smith@amd.com>

Author: rasmith

tests/lora/test_punica_ops.py

@@ -395,6 +395,7 @@ def test_kernels(
     Tests LoRA kernels.
     """
     torch.set_default_device(device)
+    torch.cuda.set_device(device)
     set_random_seed(seed)
 
     if op_type == "shrink":
@@ -447,6 +448,7 @@ def test_kernels_hidden_size(
     Tests SGMV and LoRA kernels.
     """
     torch.set_default_device(device)
+    torch.cuda.set_device(device)
     set_random_seed(seed)
 
     if op_type == "shrink":