BUG: Pull request description goes out of token

[sauravpanda]

Traceback (most recent call last):
File "/app/github_app/github_helper/pull_requests.py", line 179, in process_pr_desc
desc = reviewer.generate_pull_request_desc(
File "/root/.cache/pypoetry/virtualenvs/cloudcode-bot-9TtSrW0h-py3.9/lib/python3.9/site-packages/kaizen/reviewer/code_review.py", line 160, in generate_pull_request_desc
resp, usage = self.provider.chat_completion(prompt, user=user)
File "/root/.cache/pypoetry/virtualenvs/cloudcode-bot-9TtSrW0h-py3.9/lib/python3.9/site-packages/kaizen/llms/provider.py", line 42, in chat_completion

8:24
response = litellm.completion(messages=messages, user=user, **custom_model)
File "/root/.cache/pypoetry/virtualenvs/cloudcode-bot-9TtSrW0h-py3.9/lib/python3.9/site-packages/litellm/utils.py", line 945, in wrapper
raise e
File "/root/.cache/pypoetry/virtualenvs/cloudcode-bot-9TtSrW0h-py3.9/lib/python3.9/site-packages/litellm/utils.py", line 840, in wrapper
result = original_function(*args, **kwargs)
File "/root/.cache/pypoetry/virtualenvs/cloudcode-bot-9TtSrW0h-py3.9/lib/python3.9/site-packages/litellm/main.py", line 2523, in completion
raise exception_type(
File "/root/.cache/pypoetry/virtualenvs/cloudcode-bot-9TtSrW0h-py3.9/lib/python3.9/site-packages/litellm/utils.py", line 7000, in exception_type
raise e
File "/root/.cache/pypoetry/virtualenvs/cloudcode-bot-9TtSrW0h-py3.9/lib/python3.9/site-packages/litellm/utils.py", line 5471, in exception_type
raise ContextWindowExceededError(
litellm.exceptions.ContextWindowExceededError: litellm.BadRequestError: litellm.ContextWindowExceededError: ContextWindowExceededError: OpenAIException - Error code: 400 - {'error': {'message': "This model's maximum context length is 16385 tokens. However, your messages resulted in 37662 tokens. Please reduce the length of the messages.", 'type': 'invalid_request_error', 'param': 'messages', 'code': 'context_length_exceeded'}}