[Platform][Worker][ModelRunner] Add LoRA & Multi-LoRA support (#521)

### What this PR does / why we need it?
According to this RFC [[RFC]: Join the MultiLora and MultiLora Dynammic
Serving feature develop
#396](https://github.com/vllm-project/vllm-ascend/issues/396) and this
[vLLM Ascend Roadmap Q2 2025
#448](https://github.com/vllm-project/vllm-ascend/issues/448), we pull
request relavant code to support (1) Multi-LoRA and (2) Multi-LoRA
Dynamic Serving.

LoRA reference is here: [LoRA
reference](https://docs.vllm.ai/en/latest/features/lora.html)

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

Following openai HTTP apis will be supported:
/v1/load_lora_adapter
/v1/unload_lora_adapter

### How was this patch tested?
git clone https://github.com/vllm-project/vllm.git
cd vllm/examples/offline_inference/ && python3 multilora_inference.py

---------

Signed-off-by: paulyu <paulyu0307@gmail.com>
Co-authored-by: paulyu <paulyu0307@gmail.com>

vllm_ascend/platform.py

@@ -141,6 +141,10 @@ class NPUPlatform(Platform):
             return "vllm_ascend.attention.attention.AscendMLAAttentionBackend"
         return "vllm_ascend.attention.attention.AscendAttentionBackend"
 
+    @classmethod
+    def get_punica_wrapper(cls) -> str:
+        return "vllm_ascend.lora.punica_wrapper.punica_npu.PunicaWrapperNPU"
+
     @classmethod
     def get_current_memory_usage(cls,
                                  device: Optional[torch.types.Device] = None