[Feature]: implement the fusion of allreduce and matmul in prefill phase when tp is enabled (#1926)

### What this PR does / why we need it?
it'll execute allreduce and malmul seperately in vllm RowParallelLinear
forward funcion, this function use torch_npu.npu_mm_all_reduce_base to
execute allreduce and matmul in a fused kernel way. this will gain a 20%
performance
promotion in eager mode.
### Does this PR introduce _any_ user-facing change?
this PR introduce a new env `VLLM_ASCEND_ENABLE_MATMUL_ALLREDUCE` to
control whether enable the feature or not.

### How was this patch tested?
the patch is tested by adding a new test file `test_patch_linear.py` to
guard the ut


- vLLM version: v0.10.0
- vLLM main:
7728dd77bb

Signed-off-by: Ronald1995 <ronaldautomobile@163.com>

tests/ut/base.py

@@ -19,10 +19,6 @@ import pytest
 
 from vllm_ascend.utils import adapt_patch, register_ascend_customop
 
-# fused moe ops test will hit the infer_schema error, we need add the patch
-# here to make the test pass.
-import vllm_ascend.patch.worker.patch_common.patch_utils  # type: ignore[import]  # isort: skip  # noqa
-
 
 class TestBase(unittest.TestCase):