[BugFix] fix ep=1 etp=16

[ttanzhiqiang]

What this PR does / why we need it?

Fixed ep=1 etp=16 bug #971, Refer to #863 this pr

Does this PR introduce any user-facing change?

Added etp logic branch in deepseekv2 and fused_moe

How was this patch tested?

nohup python -m vllm.entrypoints.openai.api_server --model=/mnt/deepseek/DeepSeek-R1-W8A8-VLLM
--trust-remote-code
--distributed-executor-backend=mp
-tp=16
-dp=1
--port 8006
--max-num-seqs 24
--max-model-len 32768
--max-num-batched-tokens 32768
--block-size 128
--enable-expert-parallel
--compilation_config 0
--gpu-memory-utilization 0.96
--additional-config '{"expert_tensor_parallel_size":1, "ascend_scheduler_config":{}}' &> run.log &

nohup python -m vllm.entrypoints.openai.api_server --model=/mnt/deepseek/DeepSeek-R1-W8A8-VLLM
--trust-remote-code
--distributed-executor-backend=mp
-tp=16
-dp=1
--port 8006
--max-num-seqs 24
--max-model-len 32768
--max-num-batched-tokens 32768
--block-size 128
--enable-expert-parallel
--compilation_config 0
--gpu-memory-utilization 0.96
--additional-config '{"expert_tensor_parallel_size":16, "ascend_scheduler_config":{}}' &> run.log &

[ttanzhiqiang]

@wangxiyuan @Angazenn

[ttanzhiqiang]

The latest branch is running smoothly, vllm-ascend: commit 6eddbd2
vllm: releases/v0.9.0
nohup python -m vllm.entrypoints.openai.api_server --model=/mnt/deepseek/DeepSeek-R1-W8A8-VLLM
--trust-remote-code
--distributed-executor-backend=mp
-tp=16
-dp=1
--port 8006
--max-num-seqs 24
--max-model-len 32768
--max-num-batched-tokens 32768
--block-size 128
--enable-expert-parallel
--compilation_config 0
--gpu-memory-utilization 0.96
--additional-config '{"expert_tensor_parallel_size":1}' &> run.log &

nohup python -m vllm.entrypoints.openai.api_server --model=/mnt/deepseek/DeepSeek-R1-W8A8-VLLM \ --trust-remote-code \ --distributed-executor-backend=mp \ -tp=16 \ -dp=1 \ --port 8006 \ --max-num-seqs 24 \ --max-model-len 32768 \ --max-num-batched-tokens 32768 \ --block-size 128 \ --enable-expert-parallel \ --compilation_config 0 \ --gpu-memory-utilization 0.96 \ --additional-config '{"expert_tensor_parallel_size":16}' &> run.log &

[github-actions]

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

[ttanzhiqiang]

#1012 do it