Upgrade to 0.11.1 newest vllm commit (#3762)

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

c9461e05a4

Fix ```spec decode rejection sampler```, caused by
https://github.com/vllm-project/vllm/pull/26060
Fix some ```import```, caused by
https://github.com/vllm-project/vllm/pull/27374
Fix ```scheduler_config.send_delta_data```, caused by
https://github.com/vllm-project/vllm-ascend/pull/3719
Fix ```init_with_cudagraph_sizes```, caused by
https://github.com/vllm-project/vllm/pull/26016
Fix ```vl model```of replacing PatchEmbed's conv3d to linear layer,
caused by https://github.com/vllm-project/vllm/pull/27418

### Does this PR introduce _any_ user-facing change?
N/A

### How was this patch tested?
CI passed with new added/existing test.


- vLLM version: v0.11.0rc3
- vLLM main:
c9461e05a4

---------

Signed-off-by: Icey <1790571317@qq.com>

vllm_ascend/worker/model_runner_v1.py

@@ -72,7 +72,7 @@ from vllm.pooling_params import PoolingParams
 from vllm.sampling_params import SamplingType
 from vllm.sequence import IntermediateTensors
 from vllm.tasks import GenerationTask, PoolingTask, SupportedTask
-from vllm.utils import cdiv, is_pin_memory_available
+from vllm.utils import cdiv
 from vllm.utils.jsontree import json_map_leaves
 from vllm.v1.attention.backends.gdn_attn import GDNAttentionMetadataBuilder
 from vllm.v1.attention.backends.utils import (
@@ -159,13 +159,14 @@ else:
 if vllm_version_is("0.11.0"):
     from vllm.attention.layer import Attention
     from vllm.config import CompilationLevel
-    from vllm.utils import LazyLoader
+    from vllm.utils import LazyLoader, is_pin_memory_available
 
     from vllm_ascend.models.layers.mla import AscendMultiHeadLatentAttention
 else:
     from vllm.attention.layer import MLAAttention
     from vllm.config import CompilationMode
     from vllm.utils.import_utils import LazyLoader
+    from vllm.utils.platform_utils import is_pin_memory_available
 
 if TYPE_CHECKING:
     import xgrammar as xgr  # type: ignore[import-untyped]
@@ -386,7 +387,11 @@ class NPUModelRunner(LoRAModelRunnerMixin):
                 self.drafter = get_spec_decode_method(
                     self.speculative_config.method, self.vllm_config,
                     self.device, self)
-                self.rejection_sampler = AscendRejectionSampler()
+                if vllm_version_is("0.11.0"):
+                    self.rejection_sampler = AscendRejectionSampler()
+                else:
+                    self.rejection_sampler = AscendRejectionSampler(
+                        self.sampler)
             self.actual_seq_lengths_q = list(
                 range(self.decode_token_per_req, self.max_num_tokens + 1,
                       self.decode_token_per_req))
@@ -1885,6 +1890,9 @@ class NPUModelRunner(LoRAModelRunnerMixin):
         # TODO: Optimize the CPU -> NPU copy.
         cu_num_draft_tokens = torch.from_numpy(cu_num_draft_tokens).to(
             self.device, non_blocking=True)
+        if not vllm_version_is("0.11.0"):
+            cu_num_sampled_tokens = torch.from_numpy(cu_num_sampled_tokens).to(
+                self.device, non_blocking=True)
         logits_indices = torch.from_numpy(logits_indices).to(self.device,
                                                              non_blocking=True)
         target_logits_indices = torch.from_numpy(target_logits_indices).to(
@@ -1896,15 +1904,25 @@ class NPUModelRunner(LoRAModelRunnerMixin):
         # draft_token_indices:      [  1,   2,   3, 105, 106, 208]
         draft_token_ids = self.input_ids[logits_indices]
         draft_token_ids = draft_token_ids[target_logits_indices + 1]
-
-        metadata = SpecDecodeMetadata(
-            draft_token_ids=draft_token_ids,
-            num_draft_tokens=num_draft_tokens.tolist(),
-            cu_num_draft_tokens=cu_num_draft_tokens,
-            target_logits_indices=target_logits_indices,
-            bonus_logits_indices=bonus_logits_indices,
-            logits_indices=logits_indices,
-        )
+        if vllm_version_is("0.11.0"):
+            metadata = SpecDecodeMetadata(
+                draft_token_ids=draft_token_ids,
+                num_draft_tokens=num_draft_tokens.tolist(),
+                cu_num_draft_tokens=cu_num_draft_tokens,
+                target_logits_indices=target_logits_indices,
+                bonus_logits_indices=bonus_logits_indices,
+                logits_indices=logits_indices,
+            )
+        else:
+            metadata = SpecDecodeMetadata(
+                draft_token_ids=draft_token_ids,
+                num_draft_tokens=num_draft_tokens.tolist(),
+                cu_num_draft_tokens=cu_num_draft_tokens,
+                cu_num_sampled_tokens=cu_num_sampled_tokens,
+                target_logits_indices=target_logits_indices,
+                bonus_logits_indices=bonus_logits_indices,
+                logits_indices=logits_indices,
+            )
         return metadata
 
     def apply_grammar_bitmask(