🔻 fix: Role and System Message Handling for ChatGPT Imports

[danny-avila]

Summary

Closes #9505
Closes #9491

I fixed bugs in the ChatGPT import functionality that were causing incorrect sender assignment and broken conversation flows when encountering system messages.

• Fixed sender assignment logic to ensure user messages always maintain 'user' sender regardless of GPT model metadata
• Implemented findNonSystemParent helper function to traverse parent hierarchy and skip system messages when establishing parent-child relationships
• Added comprehensive test coverage for both fixes including edge cases with GPT-4 models and system messages interrupting conversation flows
• Updated logger import to use data-schemas module for consistency
• Ensured message graph integrity is preserved during import process even when system messages exist between user and assistant messages

Change Type

• Bug fix (non-breaking change which fixes an issue)

Testing

I added comprehensive unit tests that verify the fixes work correctly:

Test Configuration:

Sender Assignment Test: Validates that user messages retain 'user' sender and assistant messages get correct GPT-4/GPT-3.5 sender based on model, covering multiple model types (gpt-4, gpt-4o-mini, gpt-3.5-turbo).

System Message Handling Test: Creates a conversation with a system message between assistant and user messages, then verifies the parent-child relationships remain intact by checking that the second user message correctly references the first assistant message as its parent rather than being orphaned.

Checklist

• My code adheres to this project's style guidelines
• I have performed a self-review of my own code
• I have commented in any complex areas of my code
• My changes do not introduce new warnings
• I have written tests demonstrating that my changes are effective or that my feature works
• Local unit tests pass with my changes