[CI] Turn on basic correctnes tests for V1

Signed-off-by: Tyler Michael Smith <[email protected]>

tests/basic_correctness/test_basic_correctness.py

@@ -26,6 +26,15 @@
 TARGET_TEST_SUITE = os.environ.get("TARGET_TEST_SUITE", "L4")
 
 
+@pytest.fixture(autouse=True)
+def v1(run_with_both_engines):
+    # Simple autouse wrapper to run both engines for each test
+    # This can be promoted up to conftest.py to run for every
+    # test in a package
+    pass
+
+
+
 def test_vllm_gc_ed():
     """Verify vllm instance is GC'ed when it is deleted"""
     llm = LLM("facebook/opt-125m")
@@ -83,7 +92,7 @@ def test_models(
         name_1="vllm",
     )
 
-
+@pytest.mark.skip_v1
 @multi_gpu_test(num_gpus=2)
 @pytest.mark.parametrize(
     "model, distributed_executor_backend, attention_backend, "
@@ -143,6 +152,7 @@ def test_models_distributed(
     )
 
 
+@pytest.mark.skip_v1
 def test_model_with_failure(vllm_runner) -> None:
     try:
         with patch("vllm.model_executor.models.opt.OPTForCausalLM.forward",
@@ -169,6 +179,7 @@ def test_model_with_failure(vllm_runner) -> None:
         os.remove(filename)
 
 
+@pytest.mark.skip_v1
 def test_failure_with_async_out_proc(vllm_runner) -> None:
 
     filename = None