[main][refactor] Refactoring forward_context and model_runner_v1 (#1979)

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

A refactoring of forward_context and model_runner_v1, add some context
which is necessary in model inference into forward_context, and refactor
dummy_run logic, make it more reasonable.
Some details for this PR:

Add `ascend_forward_context`;
Update mc2_v2 op, and support `active_mask` param;
Update scripts in examples dir;
refactor `dummy_run` logic;
Add soc_version for A2 and A3;

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

No change at user-facing.

### How was this patch tested?


- vLLM version: v0.10.0
- vLLM main:
57c22e57f9

Signed-off-by: zzzzwwjj <1183291235@qq.com>

vllm_ascend/distributed/parallel_state.py

@@ -0,0 +1,48 @@
+from typing import Optional
+
+import torch
+from vllm.config import ParallelConfig
+from vllm.distributed.parallel_state import (GroupCoordinator, get_world_group,
+                                             init_model_parallel_group)
+
+# Currently, mc2 op need their own group coordinator.
+_MC2: Optional[GroupCoordinator] = None
+
+
+def get_mc2_group() -> GroupCoordinator:
+    assert _MC2 is not None, ("mc2 group is not initialized")
+    return _MC2
+
+
+def model_parallel_initialized():
+    return (_MC2 is not None)
+
+
+def init_ascend_model_parallel(parallel_config: ParallelConfig, ):
+    if model_parallel_initialized():
+        return
+    assert torch.distributed.is_initialized()
+    world_size = torch.distributed.get_world_size()
+    backend = torch.distributed.get_backend(get_world_group().device_group)
+
+    # The layout of all ranks: ExternalDP * EP
+    # ExternalDP is the data parallel group that is not part of the model,
+    # every dp rank can generate independently (in verl integration).
+    all_ranks = torch.arange(world_size).reshape(
+        -1, parallel_config.data_parallel_size *
+        parallel_config.tensor_parallel_size)
+    global _MC2
+    group_ranks = all_ranks.unbind(0)
+    group_ranks = [x.tolist() for x in group_ranks]
+
+    _MC2 = init_model_parallel_group(group_ranks,
+                                     get_world_group().local_rank,
+                                     backend,
+                                     group_name="mc2")
+
+
+def destroy_ascend_model_parallel():
+    global _MC2
+    if _MC2:
+        _MC2.destroy()
+    _MC2 = None