[Feature] support aclgraph for model runner v2

[Ronald1995]

What this PR does / why we need it?

This PR aims to support aclgraph for model runner v2, please see RFC #5208. The PR contains these modifications:

• adapt to newest commit of vllm main branch.
• supply a unified interface of extra forward context for both model runner v1 and model runner v2.
• implement graph mode for main model.

Does this PR introduce any user-facing change?

no

How was this patch tested?

• vLLM version: v0.16.0
• vLLM main: vllm-project/vllm@4034c3d

[realliujiaxu]

Suggestion: Simplify _ExtraForwardContextProxy

To reduce boilerplate from ExtraForwardContext.xxx() wrappers, we can use a minimal proxy with attribute-style access.

# ascend_forward_context.py - simplified proxy

class _ExtraForwardContextProxy:
    """Unified forward-context access for v1/v2 model runners."""

    @staticmethod
    def _ctx():
        return get_forward_context()

    def __getattr__(self, name: str) -> Any:
        ctx = self._ctx()
        if envs_vllm.VLLM_USE_V2_MODEL_RUNNER:
            return ctx.additional_kwargs[name]
        return getattr(ctx, name)

    def __setattr__(self, name: str, value: Any) -> None:
        ctx = self._ctx()
        if envs_vllm.VLLM_USE_V2_MODEL_RUNNER:
            ctx.additional_kwargs[name] = value
        else:
            setattr(ctx, name, value)


# usage: from vllm_ascend.ascend_forward_context import extra_ctx
extra_ctx = _ExtraForwardContextProxy()

Call comparison

Current (this PR)	With proxy
ExtraForwardContext.num_tokens()	extra_ctx.num_tokens
ExtraForwardContext.moe_comm_method()	extra_ctx.moe_comm_method
ExtraForwardContext.set_num_tokens(val)	extra_ctx.num_tokens = val
ExtraForwardContext.is_draft_model()	extra_ctx.is_draft_model

[Ronald1995]

Suggestion: Simplify _ExtraForwardContextProxy

To reduce boilerplate from ExtraForwardContext.xxx() wrappers, we can use a minimal proxy with attribute-style access.

# ascend_forward_context.py - simplified proxy

class _ExtraForwardContextProxy:
    """Unified forward-context access for v1/v2 model runners."""

    @staticmethod
    def _ctx():
        return get_forward_context()

    def __getattr__(self, name: str) -> Any:
        ctx = self._ctx()
        if envs_vllm.VLLM_USE_V2_MODEL_RUNNER:
            return ctx.additional_kwargs[name]
        return getattr(ctx, name)

    def __setattr__(self, name: str, value: Any) -> None:
        ctx = self._ctx()
        if envs_vllm.VLLM_USE_V2_MODEL_RUNNER:
            ctx.additional_kwargs[name] = value
        else:
            setattr(ctx, name, value)


# usage: from vllm_ascend.ascend_forward_context import extra_ctx
extra_ctx = _ExtraForwardContextProxy()

Call comparison

Current (this PR) With proxy
ExtraForwardContext.num_tokens() extra_ctx.num_tokens
ExtraForwardContext.moe_comm_method() extra_ctx.moe_comm_method
ExtraForwardContext.set_num_tokens(val) extra_ctx.num_tokens = val
ExtraForwardContext.is_draft_model() extra_ctx.is_draft_model

good idea, i have refined code by your suggestion.

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.