Updates to the Project for Support of sqlcoder-7b and sqlcoder2-15b

[wangzaistone]

1. Model Fine-Tuning Support: The project now supports fine-tuning for two distinct base models: sqlcoder-7b(mistral) and sqlcoder2-15b(starcoder) . This includes a complete cycle of training, prediction, and evaluation, all of which have been successfully tested.
2. Documentation Update: Relevant documentation has been updated to reflect these new changes. This includes detailed instructions and insights into the model fine-tuning process for both sqlcoder-7b and sqlcoder2-15b.
3. Updated Package Dependencies: The project’s base environment now includes updated versions of required packages.
4. The contact information for the WeChat group has been updated.
   test pass screenshot as follows :
   for sqlcoder 7b(mistral) ,0.1 epoch data , lora train ,just test

for sqlcoder15b(starcoder) ,0.1 epoch data , qlora train ,just test

[csunny]

r+