[1/N][Refactor][Quantization] remove redundant quantizer class (#2680)

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

AscendQuantizer/LLMQuantizer class is used to select quant method based
on quant config and some other arguments,
but it is more simple and clean replacing these classes with map. So i
remove them.

### Does this PR introduce _any_ user-facing change?
No 

### How was this patch tested?

ut and e2e test


- vLLM version: v0.10.1.1
- vLLM main:
6997a25ac6

Signed-off-by: 22dimensions <waitingwind@foxmail.com>

tests/ut/torchair/ops/test_torchair_fused_moe.py

@@ -24,7 +24,6 @@ from vllm.model_executor.layers.fused_moe import FusedMoEMethodBase
 
 from vllm_ascend.ascend_forward_context import _get_fused_moe_state
 from vllm_ascend.quantization.quant_config import AscendFusedMoEMethod
-from vllm_ascend.quantization.quantizer import W8A8Quantizer
 from vllm_ascend.torchair.ops.torchair_fused_moe import (
     TorchairAscendFusedMoE, TorchairAscendUnquantizedFusedMoEMethod)
 from vllm_ascend.utils import AscendSocVersion, adapt_patch  # noqa E402
@@ -236,12 +235,9 @@ class TestTorchairAscendFusedMoe:
         mock_quant_method = MockFusedMoEMethod()
         mock_quant_config.get_quant_method.return_value = mock_quant_method
         mock_quant_config.is_layer_skipped_ascend.return_value = False
-        with patch(
-                'vllm_ascend.quantization.quantizer.AscendQuantizer.get_quantizer',
-                return_value=W8A8Quantizer):
+        with patch("vllm_ascend.quantization.quant_config.get_quant_method"):
             moe = TorchairAscendFusedMoE(**default_moe_config,
                                          quant_config=mock_quant_config)
-
             assert moe.quant_method is not None
             assert isinstance(moe.quant_method, AscendFusedMoEMethod)
 

tests/ut/torchair/test_utils.py

@@ -6,7 +6,6 @@ from unittest.mock import MagicMock, patch
 import torch
 
 from tests.ut.base import TestBase
-from vllm_ascend.quantization.quantizer import SUPPORT_ASCEND_QUANTIZER_TYPE
 from vllm_ascend.torchair import utils
 
 
@@ -135,15 +134,3 @@ class TestTorchairUtils(TestBase):
 
         utils.converting_weight_acl_format(model, ACL_FORMAT_FRACTAL_NZ)
         mock_npu_cast.assert_not_called()
-
-    def test_torchair_quant_method_register(self):
-
-        TorchairW8A8DYNAMICQuantizer = SUPPORT_ASCEND_QUANTIZER_TYPE[
-            "W8A8_DYNAMIC"]
-        TorchairW4A8DYNAMICQuantizer = SUPPORT_ASCEND_QUANTIZER_TYPE[
-            "W4A8_DYNAMIC"]
-        utils.torchair_quant_method_register()
-        self.assertNotEqual(TorchairW8A8DYNAMICQuantizer,
-                            SUPPORT_ASCEND_QUANTIZER_TYPE["W8A8_DYNAMIC"])
-        self.assertNotEqual(TorchairW4A8DYNAMICQuantizer,
-                            SUPPORT_ASCEND_QUANTIZER_TYPE["W4A8_DYNAMIC"])