Drop 0.11.0 support (#4377)

There is a lot hack code for v0.11.0, which makes the code hard to
upgrade to newer vLLM version. Since v0.11.0 will release soon. Let's
drop v0.11.0 support first. Then we'll upgrade to v0.11.2 soon.


- vLLM version: v0.11.0
- vLLM main:
2918c1b49c

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

tests/e2e/conftest.py

@@ -45,6 +45,7 @@ from vllm.inputs import TextPrompt
 from vllm.outputs import RequestOutput
 from vllm.platforms import current_platform
 from vllm.transformers_utils.utils import maybe_model_redirect
+from vllm.utils.network_utils import get_open_port
 
 from tests.e2e.model_utils import (TokensTextLogprobs,
                                    TokensTextLogprobsPromptLogprobs)
@@ -54,12 +55,6 @@ from vllm_ascend.ascend_config import clear_ascend_config
 # we not explicitly patch here, some of them might be effectiveless
 # in pytest scenario
 from vllm_ascend.utils import adapt_patch  # noqa E402
-from vllm_ascend.utils import vllm_version_is
-
-if vllm_version_is("0.11.0"):
-    from vllm.utils import get_open_port
-else:
-    from vllm.utils.network_utils import get_open_port
 
 adapt_patch(True)
 adapt_patch(False)

tests/e2e/multicard/test_aclgraph_capture_replay.py

@@ -23,13 +23,7 @@ from unittest.mock import patch
 
 import pytest
 import torch
-
-from vllm_ascend.utils import vllm_version_is
-
-if vllm_version_is("0.11.0"):
-    from vllm.utils import get_open_port
-else:
-    from vllm.utils.network_utils import get_open_port
+from vllm.utils.network_utils import get_open_port
 
 MODELS = [
     "Qwen/Qwen3-0.6B",

tests/e2e/multicard/test_single_request_aclgraph.py

@@ -19,14 +19,9 @@ from typing import Any
 
 import openai
 import pytest
+from vllm.utils.network_utils import get_open_port
 
 from tests.e2e.conftest import RemoteOpenAIServer
-from vllm_ascend.utils import vllm_version_is
-
-if vllm_version_is("0.11.0"):
-    from vllm.utils import get_open_port
-else:
-    from vllm.utils.network_utils import get_open_port
 
 MODELS = ["Qwen/Qwen3-30B-A3B", "vllm-ascend/DeepSeek-V2-Lite-W8A8"]
 

tests/e2e/nightly/models/test_qwen3_32b.py

@@ -18,15 +18,10 @@ from typing import Any
 
 import openai
 import pytest
+from vllm.utils.network_utils import get_open_port
 
 from tests.e2e.conftest import RemoteOpenAIServer
 from tools.aisbench import run_aisbench_cases
-from vllm_ascend.utils import vllm_version_is
-
-if vllm_version_is("0.11.0"):
-    from vllm.utils import get_open_port
-else:
-    from vllm.utils.network_utils import get_open_port
 
 MODELS = [
     "Qwen/Qwen3-32B",

tests/e2e/singlecard/multi-modal/test_internvl.py

@@ -25,7 +25,6 @@ from vllm.assets.image import ImageAsset
 
 from tests.e2e.conftest import VllmRunner
 from tests.e2e.model_utils import check_outputs_equal
-from vllm_ascend.utils import vllm_version_is
 
 MODELS = [
     "OpenGVLab/InternVL2-8B",
@@ -34,13 +33,6 @@ MODELS = [
     "OpenGVLab/InternVL3_5-8B",
 ]
 
-# skip testing InternVL3-8B and InternVL3_5-8B on 0.11.0 due to https://github.com/vllm-project/vllm-ascend/issues/3925.
-if vllm_version_is("0.11.0"):
-    MODELS = [
-        "OpenGVLab/InternVL2-8B",
-        "OpenGVLab/InternVL2_5-8B",
-    ]
-
 
 @pytest.mark.parametrize("model", MODELS)
 def test_internvl_basic(model: str):

tests/e2e/singlecard/test_camem.py

@@ -23,16 +23,11 @@ from unittest.mock import patch
 
 import torch
 from vllm import SamplingParams
+from vllm.utils.mem_constants import GiB_bytes
 
 from tests.e2e.conftest import VllmRunner
 from tests.e2e.utils import fork_new_process_for_each_test
 from vllm_ascend.device_allocator.camem import CaMemAllocator
-from vllm_ascend.utils import vllm_version_is
-
-if vllm_version_is("0.11.0"):
-    from vllm.utils import GiB_bytes
-else:
-    from vllm.utils.mem_constants import GiB_bytes
 
 
 @fork_new_process_for_each_test