[CI/UT][bugfix] fix v0 spec decode (#1321)

### What this PR does / why we need it?
1. [PR913](https://github.com/vllm-project/vllm-ascend/pull/913)
introduced an error that caused V0's spec decode function to fail.
[PR1109](https://github.com/vllm-project/vllm-ascend/pull/1109) wanted
to fix this problem. Unfortunately, the fix broke the ngram function. I
fixed the ngram function in this PR. **PS**: Q: Why is there a problem
when ngram is not found when pr1109 is merged? A: The newly introduced
problem will only appear when tp>1, and the use cases on CI are all tp=1
2. In versions after 0.7.3, vllm-ascend deleted some spec decode UTs to
avoid CI taking too long, including eagle speculative UTs, which made CI
unable to take care of the eagle function. I added
it(`test_eagle_correctness.py`) back in this PR
3. Because of the reason mentioned in 2, the current version of Eagle
has a problem. I located and fixed this problem. It was because vllm's
`draft_model_runner.py` was changed and vllm-ascend was not synchronized
in time.
4. Currently, the UTs of v0 and v1 are mixed in the spec_decode
directory. I split them into two directories: spec_decode_v0 and
spec_decode_v1.
5. i found
`vllm.spec_decode.multi_step_worker.MultiStepWorker.set_include_gpu_probs_tensor`
and
`vllm.spec_decode.multi_step_worker.MultiStepWorker.set_should_modify_greedy_probs_inplace`
have changed in vllm, so i remove it in this pr.

### Does this PR introduce _any_ user-facing change?
This PR fixes the functions of ngram and eagle spec decode in the v0
engine

### How was this patch tested?
tested by CI

Signed-off-by: mengwei805 <mengwei25@huawei.com>

vllm_ascend/patch/__init__.py

@@ -100,18 +100,6 @@
 #    Future Plan:
 #       Revert it when the related pr is merged in vllm and vllm-ascend.
 #
-#   2. `vllm.spec_decode.multi_step_worker.MultiStepWorker.set_include_gpu_probs_tensor` and
-#       `vllm.spec_decode.multi_step_worker.MultiStepWorker.set_should_modify_greedy_probs_inplace`
-#    Why:
-#       vLLM `Remove Sampler from Model Code` so vllm-ascend needs adapt to this change.
-#    How：
-#       Use vLLM 0.8.4 method to patch it.
-#    Related PR (if no, explain why):
-#       - https://github.com/vllm-project/vllm/pull/15195
-#       - https://github.com/vllm-project/vllm-ascend/pull/395
-#    Future Plan:
-#       Remove it when we identify the reasons clearly.
-#
 # ** File: worker/patch_common/patch_spec_decode_worker.py **
 # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 #   1. `vllm.spec_decode.spec_decode_worker.SpecDecodeWorker.create_worker`

vllm_ascend/patch/worker/patch_common/patch_multi_step_worker.py

@@ -88,20 +88,4 @@ def sampler_output(
     return filtered_model_outputs, True
 
 
-def set_include_gpu_probs_tensor(self) -> None:
-    # Need include_gpu_probs_tensor for MultiSteoWorker
-    if hasattr(self.model_runner.model, "sampler"):
-        self.model_runner.model.sampler.include_gpu_probs_tensor = True
-    self.model_runner.sampler.include_gpu_probs_tensor = True
-
-
-def set_should_modify_greedy_probs_inplace(self) -> None:
-    if hasattr(self.model_runner.model, "sampler"):
-        self.model_runner.model.sampler.should_modify_greedy_probs_inplace = (
-            True)
-    self.model_runner.sampler.should_modify_greedy_probs_inplace = True
-
-
 MultiStepWorker.sampler_output = torch.inference_mode()(sampler_output)
-MultiStepWorker.set_include_gpu_probs_tensor = set_include_gpu_probs_tensor
-MultiStepWorker.set_should_modify_greedy_probs_inplace = set_should_modify_greedy_probs_inplace

vllm_ascend/patch/worker/patch_common/patch_spec_decode_worker.py

@@ -57,11 +57,6 @@ def create_worker(
     ngram_prompt_lookup_min = (
         draft_worker_kwargs.pop("ngram_prompt_lookup_min"))
 
-    # TODO(Yizhou): A quick fix, must be refactored ASAP
-    draft_worker_kwargs["vllm_config"].parallel_config.expert_parallel_size = 1
-    draft_worker_kwargs[
-        "vllm_config"].parallel_config.expert_tensor_parallel_size = 1
-
     draft_model_config = draft_worker_kwargs["vllm_config"].model_config
     draft_parallel_config: ParallelConfig = draft_worker_kwargs[
         'vllm_config'].parallel_config
@@ -72,6 +67,13 @@ def create_worker(
         proposer_worker.set_ngram_window_size(ngram_prompt_lookup_min,
                                               ngram_prompt_lookup_max)
     else:
+        # TODO(Yizhou): A quick fix, must be refactored ASAP
+        # ngram need not this fix.
+        draft_worker_kwargs[
+            "vllm_config"].parallel_config.expert_parallel_size = 1
+        draft_worker_kwargs[
+            "vllm_config"].parallel_config.expert_tensor_parallel_size = 1
+
         draft_tp = draft_parallel_config.tensor_parallel_size
         target_tp = scorer_worker.parallel_config.tensor_parallel_size
 

vllm_ascend/worker/draft_model_runner.py

@@ -51,12 +51,6 @@ class TP1DraftModelRunner(ModelRunnerWrapperBase):
     """
 
     def __init__(self, model_runner: ModelRunnerBase):
-        if hasattr(
-                model_runner,
-                "return_hidden_states") and model_runner.return_hidden_states:
-            raise ValueError(
-                "return_hidden_states is not supported for TP1DraftModelRunner."
-            )
         super().__init__(model_runner)
 
         self.indices_of_seq_with_bonus_tokens = None
@@ -211,6 +205,9 @@ class TP1DraftModelRunner(ModelRunnerWrapperBase):
             if self.prompt_adapter_config is not None:
                 raise ValueError("TP1DraftModelRunner has no support for "
                                  "prompt_adapter_config")
+            if model_input.inputs_embeds is not None:
+                raise ValueError("TP1DraftModelRunner has no support for "
+                                 "inputs_embeds")
             if model_input.multi_modal_kwargs:
                 raise ValueError(
                     "TP1DraftModelRunner has no support for multi_modal_kwargs"
@@ -272,6 +269,7 @@ class TP1DraftModelRunner(ModelRunnerWrapperBase):
 
                 hidden_states = model_executable(
                     input_ids=model_input.input_tokens,
+                    inputs_embeds=None,
                     positions=model_input.input_positions,
                     intermediate_tensors=intermediate_tensors,
                     **MultiModalKwargs.as_kwargs(multi_modal_kwargs,
@@ -293,6 +291,9 @@ class TP1DraftModelRunner(ModelRunnerWrapperBase):
             )
             outputs.append(output)
 
+            if self.return_hidden_states and is_fallback:
+                output.hidden_states = hidden_states
+
             if model_input.attn_metadata.num_prefills == 0 \
                 and self.indices_of_seq_with_bonus_tokens is not None:
                 assert output.sampled_token_ids is not None