[3/N][CI/UT] add spec decode e2e UT

Co-authored-by: mengwei805 <[email protected]>
Co-authored-by: MengqingCao <[email protected]>
Signed-off-by: mengwei805 <[email protected]>

Author: mengwei805

.github/workflows/vllm_ascend_test.yaml

@@ -148,7 +148,7 @@ jobs:
       - name: Run vllm-project/vllm-ascend key feature test
         if: steps.filter.outputs.speculative_tests_changed
         run: |
-          pytest -sv tests/spec_decode
+          pytest -sv tests/spec_decode/e2e/test_eagle_correctness.py::test_llama2_eagle_e2e_greedy_correctness
 
       - name: Run vllm-project/vllm test
         run: |

tests/conftest.py

@@ -62,14 +62,15 @@ def __init__(
         enable_chunked_prefill: bool = False,
         swap_space: int = 4,
         enforce_eager: Optional[bool] = False,
+        trust_remote_code: Optional[bool] = True,
         **kwargs,
     ) -> None:
         self.model = LLM(
             model=model_name,
             task=task,
             tokenizer=tokenizer_name,
             tokenizer_mode=tokenizer_mode,
-            trust_remote_code=True,
+            trust_remote_code=trust_remote_code,
             dtype=dtype,
             swap_space=swap_space,
             enforce_eager=enforce_eager,