inital changes into enable openai finetuning

[danikhan632]

Title: Add OpenAI Fine-Tuning Support

Summary:
This pull request adds support for OpenAI's fine-tuning API, allowing users to fine-tune models using OpenAI's ecosystem. It also integrates with the client API and improves configuration and data loading logic to support OpenAI datasets. It's pretty rough around the edges as far as features go but it works.

Key Changes:

Testing and Verification:

• Verified dataset loading for both standard and OpenAI-specific configurations.
• Tested the new openai_train command with a sample OpenAI dataset.
• Updated and tested the YAML configuration file (llama3_lora_sft.yaml) with the OpenAI dataset setup.

Next Steps:

• Further validation is needed for larger datasets and more complex scenarios.
• Expand documentation for the openai_train command and dataset configuration in dataset_info.json.