[Feat] Support native Kimi-K2-Thinking native W4A16 quantized experts weights (#4516)

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

Adds W4A16 quantization method for the Kimi-K2-Thinking model and
updates relevant modules to support the new quantization method.

- Implements complete W4A16 quantization method including weight
packing/unpacking, per-group quantization parameter generation,
post-processing logic and MoE method application.
- Adds parameters `use_int4_w4a16`, `w1_offset` and `w2_offset`, adjusts
`with_quant` conditional logic to support W4A16 matrix multiplication.
- Adds `packed_modules_model_mapping` for Kimi-K2-Thinking model and
processing logic for `weight_packed` field.

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: zhoux77899 <zhouxiang100@huawei.com>
Signed-off-by: Ruri <33858552+zhoux77899@users.noreply.github.com>
Signed-off-by: Ruri <zhouxiang100@huawei.com>

vllm_ascend/worker/model_runner_v1.py

@@ -3471,13 +3471,13 @@ class NPUModelRunner(LoRAModelRunnerMixin, ECConnectorModelRunnerMixin):
                     # as it only support the 0-dim of kv_cache is `num_blocks`.
                     # For deepseek mla, we need to spilt cache tensor accrodding to the nope head dim
                     # and rope head dim.
-                    if self.model_config.is_deepseek_mla:
+                    if self.model_config.use_mla:
                         head_size = self.model_config.hf_text_config.qk_rope_head_dim + \
                             self.model_config.hf_text_config.kv_lora_rank
 
                     dsa_k_cache_factor = None
                     dsa_k_cache_size = None
-                    if not self.model_config.is_deepseek_mla:
+                    if not self.model_config.use_mla:
                         # for non-mla model, use FullAttentionSpec
                         k_tensor_split_factor = 2
                         v_tensor_split_factor = 2
@@ -3627,7 +3627,7 @@ class NPUModelRunner(LoRAModelRunnerMixin, ECConnectorModelRunnerMixin):
                             kv_cache_spec.num_kv_heads,
                             kv_cache_spec.head_size)
                     dtype = kv_cache_spec.dtype
-                    if not self.model_config.is_deepseek_mla:
+                    if not self.model_config.use_mla:
                         k_shape = kv_cache_shape[1:]
                         v_shape = k_shape
                     else: