[Build/CI] Extending the set of AMD tests with Regression, Basic Correctness, Distributed, Engine, Llava Tests (#4797)

tests/engine/test_stop_reason.py

@@ -32,6 +32,7 @@ def test_stop_reason(vllm_model, example_prompts):
     # test stop token
     outputs = llm.generate(example_prompts,
                            sampling_params=SamplingParams(
+                               ignore_eos=True,
                                seed=SEED,
                                max_tokens=MAX_TOKENS,
                                stop_token_ids=[stop_token_id]))
@@ -43,7 +44,10 @@ def test_stop_reason(vllm_model, example_prompts):
     # test stop string
     outputs = llm.generate(example_prompts,
                            sampling_params=SamplingParams(
-                               seed=SEED, max_tokens=MAX_TOKENS, stop="."))
+                               ignore_eos=True,
+                               seed=SEED,
+                               max_tokens=MAX_TOKENS,
+                               stop="."))
     for output in outputs:
         output = output.outputs[0]
         assert output.finish_reason == "stop"