[Fix] Avoid AsyncEngine running the same session id

[AllentDan]

No matter how the client uses session_id, keep the sever working.

import os
from random import randint
import sys
from tqdm import tqdm

from concurrent.futures import ThreadPoolExecutor

from lmdeploy.serve.openai.api_client import APIClient

questions = ['你是谁'] * 1000

num_parallel = 256

def process_one(question, url='0.0.0.0', port='23333'):
    client = APIClient('http://{}:{}'.format(url, port))
    model_name = client.available_models[0]

    msg = [dict(role='user', content=question)]

    data = client.chat_interactive_v1(msg, session_id=randint(1, 100), repetition_penalty=1.02)
    for item in data:
        pass

    data = client.chat_completions_v1(model=model_name, messages=msg, repetition_penalty=1.02)

    for item in data:
        response = item
    
    return response

with ThreadPoolExecutor(max_workers=num_parallel) as executor:
    for response in tqdm(executor.map(process_one, questions)):
        print(response)