MistralAI broken with 1.4.0

[x] This is actually a bug report.
[ ] I am not getting good LLM Results
[ ] I have tried asking for help in the community on discord or discussions and have not received a response.
[ ] I have tried searching the documentation and have not found an answer.

What Model are you using?

[ ] gpt-3.5-turbo
[ ] gpt-4-turbo
[ ] gpt-4
[x] Mistral large latest

Describe the bug A clear and concise description of what the bug is.

Instructor works with MistralAI in 1.3.7, but not 1.4.0.

Failed after retries: <bound method Future.exception of <Future at 0x113f5eec0 state=finished raised AttributeError>>
Traceback (most recent call last):
  File "/Users/ahuang/miniconda3/envs/lumen/lib/python3.10/site-packages/instructor/retry.py", line 242, in retry_async
    return await process_response_async(
  File "/Users/ahuang/miniconda3/envs/lumen/lib/python3.10/site-packages/instructor/process_response.py", line 77, in process_response_async
    model = response_model.from_response(
  File "/Users/ahuang/miniconda3/envs/lumen/lib/python3.10/site-packages/instructor/function_calls.py", line 149, in from_response
    return cls.parse_tools(completion, validation_context, strict)
  File "/Users/ahuang/miniconda3/envs/lumen/lib/python3.10/site-packages/instructor/function_calls.py", line 318, in parse_tools
    message.refusal is None
  File "/Users/ahuang/miniconda3/envs/lumen/lib/python3.10/site-packages/pydantic/main.py", line 828, in __getattr__
    raise AttributeError(f'{type(self).__name__!r} object has no attribute {item!r}')
AttributeError: 'AssistantMessage' object has no attribute 'refusal'

To Reproduce Steps to reproduce the behavior, including code snippets of the model and the input data and openai response.

import os
import instructor

from pydantic import BaseModel
from mistralai import Mistral

# enables `response_model` in chat call
client = Mistral(api_key=os.getenv("MISTRAL_API_KEY"))

patched_chat = instructor.patch(create=client.chat.complete_async, mode=instructor.Mode.MISTRAL_TOOLS)

class UserDetails(BaseModel):
    name: str
    age: int

resp = await patched_chat(
    model="mistral-large-latest",
    response_model=UserDetails,
    messages=[
        {
            "role": "user",
            "content": f'Extract the following entities: "Jason is 20"',
        },
    ],
)
print(resp)

Expected behavior A clear and concise description of what you expected to happen.

Not crash

Screenshots If applicable, add screenshots to help explain your problem.

jxnl / instructor

MistralAI broken with 1.4.0 #968