multi-LoRA as extra models in OpenAI server (#2775)

how to serve the loras (mimicking the [multilora inference example](https://github.com/vllm-project/vllm/blob/main/examples/multilora_inference.py)):
```terminal
$ export LORA_PATH=~/.cache/huggingface/hub/models--yard1--llama-2-7b-sql-lora-test/
$ python -m vllm.entrypoints.api_server \
 --model meta-llama/Llama-2-7b-hf \
 --enable-lora \
 --lora-modules sql-lora=$LORA_PATH sql-lora2=$LORA_PATH
```
the above server will list 3 separate values if the user queries `/models`: one for the base served model, and one each for the specified lora modules. in this case sql-lora and sql-lora2 point to the same underlying lora, but this need not be the case. lora config values take the same values they do in EngineArgs

no work has been done here to scope client permissions to specific models

vllm/entrypoints/openai/api_server.py

@@ -23,6 +23,7 @@ from vllm.entrypoints.openai.protocol import CompletionRequest, ChatCompletionRe
 from vllm.logger import init_logger
 from vllm.entrypoints.openai.serving_chat import OpenAIServingChat
 from vllm.entrypoints.openai.serving_completion import OpenAIServingCompletion
+from vllm.entrypoints.openai.serving_engine import LoRA
 
 TIMEOUT_KEEP_ALIVE = 5  # seconds
 
@@ -48,6 +49,16 @@ async def lifespan(app: fastapi.FastAPI):
 app = fastapi.FastAPI(lifespan=lifespan)
 
 
+class LoRAParserAction(argparse.Action):
+
+    def __call__(self, parser, namespace, values, option_string=None):
+        lora_list = []
+        for item in values:
+            name, path = item.split('=')
+            lora_list.append(LoRA(name, path))
+        setattr(namespace, self.dest, lora_list)
+
+
 def parse_args():
     parser = argparse.ArgumentParser(
         description="vLLM OpenAI-Compatible RESTful API server.")
@@ -81,6 +92,15 @@ def parse_args():
                         help="The model name used in the API. If not "
                         "specified, the model name will be the same as "
                         "the huggingface name.")
+    parser.add_argument(
+        "--lora-modules",
+        type=str,
+        default=None,
+        nargs='+',
+        action=LoRAParserAction,
+        help=
+        "LoRA module configurations in the format name=path. Multiple modules can be specified."
+    )
     parser.add_argument("--chat-template",
                         type=str,
                         default=None,
@@ -217,8 +237,10 @@ if __name__ == "__main__":
     engine = AsyncLLMEngine.from_engine_args(engine_args)
     openai_serving_chat = OpenAIServingChat(engine, served_model,
                                             args.response_role,
+                                            args.lora_modules,
                                             args.chat_template)
-    openai_serving_completion = OpenAIServingCompletion(engine, served_model)
+    openai_serving_completion = OpenAIServingCompletion(
+        engine, served_model, args.lora_modules)
 
     # Register labels for metrics
     add_global_metrics_labels(model_name=engine_args.model)