[Feature Request] `schemas` subpackage to support different structured output schemas about camel HOT 1 OPEN

I think using Function_call to satisfy with this feature, and I show a brief example about using json as a output format.
the user should input:

input_info = [
    "Determines the answer based on their beliefs, desires, intention",
    {
        "Belief": "The belief factor.", 
        "Desire": "The desire factor.",
        "Intention": "The intention factor.",
    },
]

the first element in the input is the description when should the model struct output and the second dict is the detail about the information the user need in the response.
To fit with the function call function already in camel. We should transfer the user input to a function:

def generate_text(Believe, Desire, Intention):
    """
    Determines the answer based on their beliefs, desires, intention
    Args:
        Believe (any): The belief factor.
        Desire (any): The desire factor.
        Intention (any): The intention factor.
    Returns:
        Dict[str, Any]: A dictionary containing the model's answer with keys for Believe, Desire, Intention
    """
    model_answer = {
        "Believe": Believe,
        "Desire": Desire,
        "Intention": Intention,
    }
    return model_answer

The reason why it can structure the model output is the return of the function is an output the user want. To achieve this transfomation. I already design a function to transfer from the user input to this type function.
Finally we can generate the OpenAIFunction class using

func_call = OpenAIFunction(
    func=generate_text,
)

So I will test if this way can fit with nowaday's code, and I'm also concern about which part in repo should I put my code in.
Thanks for any comment~

from camel.