[Cleanup] Remove dead code make_attention_mask function (#5818)

### What this PR does / why we need it?

This PR removes the unused `make_attention_mask` function from
`vllm_ascend/worker/v2/attn_utils.py`.

**Why it's dead code:**
- After PR #4870 (attention mask unification refactor), attention mask
generation has been centralized in the `AttentionMaskBuilder` singleton
class
- The mask is now generated directly by metadata builders when needed
(e.g., `AscendAttentionMetadataBuilder`, `AscendMLAMetadataBuilder`)
- The `make_attention_mask` function is no longer called anywhere in the
codebase
- The function's parameters (including `attn_mask` and `spec_attn_mask`)
were also removed from `build_attn_metadata` in the same refactor

**Changes:**
- Remove `make_attention_mask` function (24 lines) from
`vllm_ascend/worker/v2/attn_utils.py`

### Does this PR introduce _any_ user-facing change?

No. This is a code cleanup that removes dead code. No user-facing
behavior changes.

### How was this patch tested?

- Verified that `make_attention_mask` is not called anywhere in the
codebase (via `grep`)
- CI tests pass to ensure no regressions
- The function has been unused since PR #4870 was merged
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef

Signed-off-by: lico67373 <918688502@qq.com>
Co-authored-by: weijinqian0 <1184188277@qq.com>

vllm_ascend/worker/v2/attn_utils.py

@@ -23,7 +23,6 @@ from typing import Any, Tuple
 import numpy as np
 import torch
 from vllm.config import VllmConfig
-from vllm.config.model import ModelDType
 from vllm.v1.attention.backends.utils import AttentionMetadataBuilder
 from vllm.v1.kv_cache_interface import EncoderOnlyAttentionSpec, KVCacheConfig
 
@@ -145,26 +144,3 @@ def build_attn_state(
     else:
         attn_state = AscendAttentionState.PrefillCacheHit
     return attn_state
-
-
-def make_attention_mask(
-    vllm_config: VllmConfig,
-    attn_state: AscendAttentionState,
-    dtype: ModelDType | torch.dtype,
-    device: torch.device,
-) -> torch.Tensor:
-    """make attention mask for npu's attention backend."""
-    attn_mask_builder = get_attn_mask_builder(device)
-    # pcp situation.
-    if attn_mask_builder is None:
-        raise ValueError("Attn mask builder is None")
-    # Pooling situation.
-    if vllm_config.model_config.runner_type == "pooling":
-        return attn_mask_builder.get_attn_mask(2048, torch.bool)
-
-    # TODO(Ronald1995) cosidering pcp.
-    if vllm_config.model_config.use_mla:
-        # mla prefill
-        if attn_state != AscendAttentionState.DecodeOnly:
-            return attn_mask_builder.get_mla_mask(dtype)
-    return attn_mask_builder.get_splitfuse_attn_mask()