Remove VLLM_USE_V1 (#4086)

Drop VLLM_USE_V1 usage.  This env has been removed from vLLM already.

- vLLM version: v0.11.0
- vLLM main:
83f478bb19

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>

tests/e2e/nightly/features/test_prefix_cache_deepseek_r1_0528_w8a8.py

@@ -72,7 +72,6 @@ async def test_models(model: str) -> None:
         "OMP_PROC_BIND": "false",
         "HCCL_BUFFSIZE": "1024",
         "PYTORCH_NPU_ALLOC_CONF": "expandable_segments:True",
-        "VLLM_USE_V1": "1"
     }
     additional_config = {
         "ascend_scheduler_config": {

tests/e2e/nightly/models/test_qwq_32b.py

@@ -73,7 +73,6 @@ async def test_models(model: str, mode: str, tp_size: int) -> None:
     env_dict = {
         "TASK_QUEUE_ENABLE": "1",
         "OMP_PROC_BIND": "false",
-        "VLLM_USE_V1": "1",
         "HCCL_OP_EXPANSION_MODE": "AIV",
         "VLLM_ASCEND_ENABLE_FLASHCOMM": "1",
         "VLLM_ASCEND_ENABLE_TOPK_OPTIMIZE": "1",

tests/e2e/pd_disaggreate/run_edge_case_test.sh

@@ -2,7 +2,6 @@
 export LCCL_DETERMINISTIC=1
 export HCCL_DETERMINISTIC=true
 export CLOSE_MATMUL_K_SHIFT=1
-export VLLM_USE_V1=1
 
 set -xe
 

tests/e2e/singlecard/test_aclgraph_mem.py

@@ -30,8 +30,6 @@ from vllm_ascend.worker.model_runner_v1 import NPUModelRunner
 MODELS = ["Qwen/Qwen3-0.6B", "vllm-ascend/DeepSeek-V2-Lite-W8A8"]
 
 
-@pytest.mark.skipif(os.getenv("VLLM_USE_V1") == "0",
-                    reason="aclgraph only support on v1")
 @pytest.mark.parametrize("model", MODELS)
 @pytest.mark.parametrize("max_tokens", [4])
 @patch.dict(os.environ, {"VLLM_ASCEND_FLASHCOMM2_PARALLEL_SIZE": "0"})