[V1][ModelRunner] Support pooling model for v1 engine (#1359)

### What this PR does / why we need it?
Change as little existing code as possible to add v1 pooling task's
support, notice that i move down the `vllm.v1.worker.gpu_input_batch` to
vllm-ascend, Considering the frequent changes in upstream interfaces, in
order to decouple, so i move it here
### How was this patch tested?
CI passed with new added/existing test, and I have a simple test was
first conducted locally which is adapted from
https://www.modelscope.cn/models/Qwen/Qwen3-Embedding-0.6B, just like
bellow：
```python
import os

import torch
from vllm import LLM


os.environ["VLLM_USE_MODELSCOPE"]="True"

def get_detailed_instruct(task_description: str, query: str) -> str:
    return f'Instruct: {task_description}\nQuery:{query}'

# Each query must come with a one-sentence instruction that describes the task
task = 'Given a web search query, retrieve relevant passages that answer the query'

queries = [
    get_detailed_instruct(task, 'What is the capital of China?'),
    get_detailed_instruct(task, 'Explain gravity')
]
# No need to add instruction for retrieval documents
documents = [
    "The capital of China is Beijing.",
    "Gravity is a force that attracts two bodies towards each other. It gives weight to physical objects and is responsible for the movement of planets around the sun."
]
input_texts = queries + documents

model = LLM(model="Qwen/Qwen3-Embedding-0.6B", task="embed")

outputs = model.embed(input_texts)
embeddings = torch.tensor([o.outputs.embedding for o in outputs])
scores = (embeddings[:2] @ embeddings[2:].T)
print(scores.tolist())
# [[0.7620252966880798, 0.14078938961029053], [0.1358368694782257, 0.6013815999031067]]
```
---------

Signed-off-by: wangli <wangli858794774@gmail.com>
Signed-off-by: wangli <858794774@qq.com>
Co-authored-by: wangli <858794774@qq.com>

tests/utils.py

@@ -23,10 +23,13 @@ import signal
 import subprocess
 import sys
 import time
+from collections.abc import Sequence
 from typing import Callable, Optional
 
 import openai
 import requests
+import torch
+import torch.nn.functional as F
 from typing_extensions import ParamSpec
 from vllm.engine.arg_utils import AsyncEngineArgs
 from vllm.entrypoints.openai.cli_args import make_arg_parser
@@ -197,3 +200,37 @@ def fork_new_process_for_each_test(
                                     f" args {args} and kwargs {kwargs}")
 
     return wrapper
+
+
+def matryoshka_fy(tensor: torch.Tensor, dimensions: int):
+    tensor = torch.tensor(tensor)
+    tensor = tensor[..., :dimensions]
+    tensor = F.normalize(tensor, p=2, dim=1)
+    return tensor
+
+
+def check_embeddings_close(
+    *,
+    embeddings_0_lst: Sequence[list[float]],
+    embeddings_1_lst: Sequence[list[float]],
+    name_0: str,
+    name_1: str,
+    tol: float = 1e-3,
+) -> None:
+    assert len(embeddings_0_lst) == len(embeddings_1_lst)
+
+    for prompt_idx, (embeddings_0, embeddings_1) in enumerate(
+            zip(embeddings_0_lst, embeddings_1_lst)):
+        assert len(embeddings_0) == len(embeddings_1), (
+            f"Length mismatch: {len(embeddings_0)} vs. {len(embeddings_1)}")
+
+        sim = F.cosine_similarity(torch.tensor(embeddings_0),
+                                  torch.tensor(embeddings_1),
+                                  dim=0)
+
+        fail_msg = (f"Test{prompt_idx}:"
+                    f"\nCosine similarity: \t{sim:.4f}"
+                    f"\n{name_0}:\t{embeddings_0[:16]!r}"
+                    f"\n{name_1}:\t{embeddings_1[:16]!r}")
+
+        assert sim >= 1 - tol, fail_msg