fix: handle context length too large

[johnwalz97]

Internal Notes for Reviewers

Hotfix for issue with context length exceeded errors.

External Release Notes

[cachafla]

Great 👌

[github-actions]

PR Summary

This pull request introduces a new function _truncate_summary in the validmind/ai/test_descriptions.py file to handle the truncation of summaries that exceed a specified maximum token length. The function uses the tiktoken library to encode the summary and ensure it does not exceed the max_tokens limit, which is set to 100,000 by default. If the summary exceeds this limit, it is truncated, and a warning is logged to indicate that the generated description may be inaccurate.

Additionally, the generate_description function has been updated to use _truncate_summary for the summary field, ensuring that summaries are appropriately truncated before being returned.

The error handling in the wrapped function has been improved to provide more specific warnings when a test result is too large to generate a description, and a default description is used instead.

Minor import reordering has been done in validmind/tests/run.py to adhere to standard import ordering practices.

Test Suggestions

• Test the _truncate_summary function with summaries of varying lengths to ensure it correctly truncates when necessary.
• Verify that the generate_description function integrates _truncate_summary correctly and returns truncated summaries when appropriate.
• Check the logging output to ensure that warnings are logged correctly when summaries are truncated or when default descriptions are used.
• Test the error handling in the wrapped function to ensure it logs the correct warnings for different types of exceptions.