[Hardware][Intel] Add CPU inference backend (#3634)

Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
Co-authored-by: Yuan Zhou <yuan.zhou@intel.com>

vllm/engine/llm_engine.py

@@ -178,6 +178,9 @@ class LLMEngine:
         if device_config.device_type == "neuron":
             from vllm.executor.neuron_executor import NeuronExecutor
             executor_class = NeuronExecutor
+        elif device_config.device_type == "cpu":
+            from vllm.executor.cpu_executor import CPUExecutor
+            executor_class = CPUExecutor
         elif parallel_config.worker_use_ray:
             initialize_ray_cluster(parallel_config)
             from vllm.executor.ray_gpu_executor import RayGPUExecutor