Updated qlora.py to fix freezing of embedding layers

[ffohturk]

Updated the way embeddings are frozen by changing the order of operations. In the original codebase, the model was loaded, LoRA-fied and then the tokenizer was resized. The resizing resets the gradients of the embedding layers and puts them to default (i.e. True). This is not what you want and to fix this, you can simple do a different order: First load the base model, then resize tokenizer, then do prepare_for_kbit_training to freeze the model's original weights, then LoRA-fy and then train.

[artidoro]

@ffohturk Thanks for helping streamline the code! This makes more sense and will make it easier for people to understand what is going on. I verified that the embeddings are indeed frozen before training.