当启用 stream=true 的时候，以流响应，返回的数据大体如下：

data: {"id":"chatcmpl-6r3B875xFqmzK9lMm8sousVO3iBN4","object":"chat.completion.chunk","created":1678101622,"model":"gpt-3.5-turbo-0301","choices":[{"delta":{"role":"assistant"},"index":0,"finish_reason":nul
l}]}

data: {"id":"chatcmpl-6r3B875xFqmzK9lMm8sousVO3iBN4","object":"chat.completion.chunk","created":1678101622,"model":"gpt-3.5-turbo-0301","choices":[{"delta":{"content":"\n\n"},"index":0,"finish_reason":null}
]}

data: {"id":"chatcmpl-6r3B875xFqmzK9lMm8sousVO3iBN4","object":"chat.completion.chunk","created":1678101622,"model":"gpt-3.5-turbo-0301","choices":[{"delta":{"content":"作"},"index":0,"finish_reason":null}]}

data: {"id":"chatcmpl-6r3B875xFqmzK9lMm8sousVO3iBN4","object":"chat.completion.chunk","created":1678101622,"model":"gpt-3.5-turbo-0301","choices":[{"delta":{"content":"为"},"index":0,"finish_reason":null}]}

data: {"id":"chatcmpl-6r3B875xFqmzK9lMm8sousVO3iBN4","object":"chat.completion.chunk","created":1678101622,"model":"gpt-3.5-turbo-0301","choices":[{"delta":{"content":"一个"},"index":0,"finish_reason":null}}

...

data: [DONE]

每一个 event data 的 json 串，得到的 content 内容仅仅是一个字，看完真觉得浪费流量啊，是否我的调用方式不对呢？有其他的参数或调用方式来避免这种浪费吗？