[refact] unified soc_version code (#4359)

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

Currently, there are two paths to judge the chip type in code,
`get_ascend_soc_version` use `get_soc_version` api in torch_npu, and
`is_310p` `use _build_info.__soc_version__`, which generate when
install. We need to unify the two paths.

We need to unify these codes based on the following points:

1. We need to ensure consistency in chip type judgment between compiling
and running states;
2. In compiling state, we need chip type to complete op's compilation,
but in running state, we only need device
type(910B/910_93/310P/910_95/etc) to make code branch judgement;
3. In compiling state, torch_npu may not have been installed yet, so we
can't use torch_npu's api.

Based on the above points, we have made the following changes:

1. When user set env `SOC_VERSION`, use it; when not set, query
soc_version by `npu-smi`;
2. generate device_type based on soc_version when compiling, and write
`__device_type__` instead of `__soc_version__` in `_build_info.py`;
3. In running state, use `__device_type__` to judge code branch.

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

When not set env `SOC_VERSION`, it will not be `ASCEND910B1` by default,
we will query soc_version by `npu-smi`. And env `SOC_VERSION` must be in
the list `soc_to_device` in `setup.py`.

- vLLM version: v0.11.0
- vLLM main:
2918c1b49c

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

tests/ut/torchair/ops/test_torchair_fused_moe.py

@@ -28,7 +28,7 @@ from vllm_ascend.quantization.quant_config import AscendFusedMoEMethod
 from vllm_ascend.torchair.ops.torchair_fused_moe import (
     TorchairAscendFusedMoE, TorchairAscendUnquantizedFusedMoEMethod)
 from vllm_ascend.utils import adapt_patch  # noqa E402
-from vllm_ascend.utils import AscendSocVersion
+from vllm_ascend.utils import AscendDeviceType
 
 adapt_patch(True)
 
@@ -398,7 +398,7 @@ class TestTorchairAscendUnquantizedFusedMoEMethod:
         forward_context = MagicMock(
             fused_moe_state=get_fused_moe_state(ep_size, is_prefill, True))
         with patch("vllm_ascend.torchair.ops.torchair_fused_moe.get_forward_context", return_value=forward_context), \
-             patch("vllm_ascend.torchair.ops.torchair_fused_moe.get_ascend_soc_version", return_value=AscendSocVersion.A3):
+             patch("vllm_ascend.torchair.ops.torchair_fused_moe.get_ascend_device_type", return_value=AscendDeviceType._910_93):
             expert_map = torch.tensor([0, 1, 2, -1, -1, -1, -1, -1])
             moe_method.ep_size = ep_size
             x = torch.randn(8, 2, 2)

tests/ut/torchair/ops/test_torchair_rotary_embedding.py

@@ -8,6 +8,7 @@ from vllm_ascend.torchair.ops.torchair_rotary_embedding import (
     _set_cos_sin_cache, custom_rotary_embedding_enabled,
     native_rope_deepseek_forward, rope_forward_oot, rotate_half,
     yarn_find_correction_dim, yarn_get_mscale)
+from vllm_ascend.utils import AscendDeviceType
 
 
 class TestCustomRotaryEmbeddingEnabled(TestBase):
@@ -107,14 +108,15 @@ class TestRopeForwardOot(TestBase):
     @patch('torch.ops._C_ascend')
     @patch(
         'vllm_ascend.torchair.ops.torchair_rotary_embedding.get_ascend_config')
-    @patch('vllm_ascend.torchair.ops.torchair_rotary_embedding.is_310p',
-           return_value=False)
+    @patch('vllm_ascend.utils.get_ascend_device_type',
+           return_value=AscendDeviceType._910_93)
     @patch(
         'vllm_ascend.torchair.ops.torchair_rotary_embedding.custom_rotary_embedding_enabled',
         return_value=True)
     @patch('torch.ops._npu_rotary_embedding')
     def test_rope_forward_oot_custom_kernel(self, mock_rotary_embedding,
-                                            mock_custom_enabled, mock_is_310p,
+                                            mock_custom_enabled,
+                                            mock_soc_version,
                                             mock_get_ascend_config, mock__c):
         mock_config = MagicMock()
         mock_config.torchair_graph_config.enabled = False

tests/ut/torchair/quantization/test_torchair_w8a8_dynamic.py

@@ -5,7 +5,7 @@ import torch
 from tests.ut.base import TestBase
 from vllm_ascend.torchair.quantization.torchair_w8a8_dynamic import (
     torchair_fused_experts_with_all2all, torchair_fused_experts_with_mc2)
-from vllm_ascend.utils import AscendSocVersion
+from vllm_ascend.utils import AscendDeviceType
 
 
 class TestAscendW8A8FusedMoEMethod(TestBase):
@@ -79,7 +79,7 @@ class TestAscendW8A8FusedMoEMethod(TestBase):
         'HCCL_INTRA_PCIE_ENABLE': '1'
     })
     @patch(
-        "vllm_ascend.torchair.quantization.torchair_w8a8_dynamic.get_ascend_soc_version"
+        "vllm_ascend.torchair.quantization.torchair_w8a8_dynamic.get_ascend_device_type"
     )
     @patch(
         'vllm_ascend.torchair.quantization.torchair_w8a8_dynamic.get_mc2_group'
@@ -94,7 +94,7 @@ class TestAscendW8A8FusedMoEMethod(TestBase):
             mock_ascend_soc_version):
         """Test expert_scales is passed in A2 SOC version with mc2 optimization"""
         # Setup mocks
-        mock_ascend_soc_version.return_value = AscendSocVersion.A2
+        mock_ascend_soc_version.return_value = AscendDeviceType._910B
 
         mock_group = MagicMock()
         mock_group.rank_in_group = 0