[Refactor] 2/N Unify all mask generation methods and cache mask (#4779)

RFC: https://github.com/vllm-project/vllm-ascend/issues/4629

Reason:

There are various types of masks here, and some of them do not have a
caching mechanism. As a result, the masks need to be initialized for
each layer, leading to waste of video memory.

At the same time, we hope to standardize the management and usage of
masks.

So we have gathered all the masks into the AttentionMaskBuilder class.

Todo:
1. remove spec_attn_mask;  @LICO1314
2. remove pcp_prefill_mask; @LICO1314


- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
Signed-off-by: weijinqian_v1 <weijinqian@huawei.com>
Signed-off-by: ZYang6263 <zy626375@gmail.com>
Signed-off-by: ZYang6263 <50876451+ZYang6263@users.noreply.github.com>
Signed-off-by: daishixun <dsxsteven@sina.com>
Signed-off-by: lulina <lina.lulina@huawei.com>
Signed-off-by: zengran <zengran2@huawei.com>
Signed-off-by: shiro-zzzz <zhangdianhao@huawei.com>
Signed-off-by: dependabot[bot] <support@github.com>
Signed-off-by: 李少鹏 <lishaopeng21@huawei.com>
Signed-off-by: xuyexiong <xuyexiong@huawei.com>
Signed-off-by: MengqingCao <cmq0113@163.com>
Signed-off-by: lhp-deep <liuhaopeng1@huawei.com>
Signed-off-by: gcanlin <canlinguosdu@gmail.com>
Signed-off-by: wangli <wangli858794774@gmail.com>
Co-authored-by: wangxiyuan <wangxiyuan1007@gmail.com>
Co-authored-by: Mengqing Cao <cmq0113@163.com>
Co-authored-by: weijinqian_v1 <weijinqian@huawei.com>
Co-authored-by: ZYang6263 <50876451+ZYang6263@users.noreply.github.com>
Co-authored-by: dsxsteven <36877507+dsxsteven@users.noreply.github.com>
Co-authored-by: LuLina <lina.lulina@huawei.com>
Co-authored-by: zengzengran <zengran2@huawei.com>
Co-authored-by: shiro-zzzz <zhangdianhao@huawei.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: shaopeng-666 <lishaopeng21@huawei.com>
Co-authored-by: xuyexiong <xuyexiong@huawei.com>
Co-authored-by: lhp-deep <liuhaopeng1@huawei.com>
Co-authored-by: Canlin Guo <canlinguosdu@gmail.com>
Co-authored-by: Li Wang <wangli858794774@gmail.com>

vllm_ascend/spec_decode/eagle_proposer.py

@@ -77,9 +77,7 @@ class EagleProposer(Proposer):
                                    1,
                                    device=device,
                                    dtype=torch.int32)
-        attn_mask_len = self.vllm_config.model_config.max_model_len
-        self.attn_mask_builder = AttentionMaskBuilder(
-            attn_mask_len, self.vllm_config.model_config.dtype, device=device)
+        self.attn_mask_builder = AttentionMaskBuilder(self.device)
 
     def load_model(self, model: nn.Module) -> None:
         target_attn_layer_names = set(
@@ -570,9 +568,7 @@ class EagleProposer(Proposer):
             self.input_ids[:batch_size] = input_ids
             self.positions[:batch_size] = clamped_positions
             self.hidden_states[:batch_size] = hidden_states
-            attn_mask = self.attn_mask_builder.get_splitfuse_attn_mask(
-                attn_metadata.seq_lens, positions_cpu,
-                self.vllm_config.model_config.dtype, self.device)
+            attn_mask = self.attn_mask_builder.get_splitfuse_attn_mask()
 
             attn_metadata.attn_mask = attn_mask
             attn_metadata.block_tables = block_table.to(device)