[Feat] Unquantized Linear to nz and control all nz-cast (#3356)

### What this PR does / why we need it?
Currently, when executing to the Linear layer of models in vLLM-Ascend,
the weights format is ND in unquantized case and skipped ascend case.
This PR supplements the execution logic for Linear layer. We use a new
global variable: VLLM_ASCEND_ENABLE_NZ. When VLLM_ASCEND_ENABLE_NZ=1 and
CANN version is 8.3, the weights of the Linear layer will be converted
to FRACTAL_NZ, in both unquantized case and skipped ascend case. We also
use VLLM_ASCEND_ENABLE_NZ to control the existing NZ conversion, such as
w8a8-quantized case.

### Does this PR introduce _any_ user-facing change?
Add a new global variable VLLM_ASCEND_ENABLE_NZ. If you want to use NZ
format, you should set VLLM_ASCEND_ENABLE_NZ=1.

### How was this patch tested?

- vLLM version: v0.11.0rc3
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

Signed-off-by: anon189Ty <Stari_Falcon@outlook.com>

vllm_ascend/torchair/utils.py

@@ -14,6 +14,7 @@ try:
 except ImportError:
     from torchair.ops import NpuStreamSwitch as _npu_stream_switch
     from torchair.ops import npu_wait_tensor as _npu_wait_tensor
+from vllm_ascend.utils import ACL_FORMAT_FRACTAL_NZ, is_enable_nz
 
 KV_CACHE_BYTES_CACHE_PATH_NAME = ".kv_cache_bytes"
 KV_CACHE_BYTES_CACHE_FILE_NAME = "kv_cache_bytes"
@@ -141,6 +142,9 @@ def converting_weight_acl_format(model, format):
         if isinstance(module, FusedMoE):
             if torch_npu.get_npu_format(module.w13_weight.data) == format:
                 return
+            if format == ACL_FORMAT_FRACTAL_NZ \
+                    and not is_enable_nz():
+                return
             module.w13_weight.data = torch_npu.npu_format_cast(
                 module.w13_weight.data, format)
             module.w2_weight.data = torch_npu.npu_format_cast(