[Misc][ModelScope] Change to use runtime VLLM_USE_MODELSCOPE (#18655)

Signed-off-by: Mengqing Cao <cmq0113@163.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <2037008807@qq.com>

tests/test_regression.py

@@ -60,6 +60,9 @@ def test_model_from_modelscope(monkeypatch: pytest.MonkeyPatch):
     # model: https://modelscope.cn/models/qwen/Qwen1.5-0.5B-Chat/summary
     with monkeypatch.context() as m:
         m.setenv("VLLM_USE_MODELSCOPE", "True")
+        # Don't use HF_TOKEN for ModelScope repos, otherwise it will fail
+        # with 400 Client Error: Bad Request.
+        m.setenv("HF_TOKEN", "")
         llm = LLM(model="qwen/Qwen1.5-0.5B-Chat")
 
         prompts = [