Add Function Calling Fine-tuning LLMs on xLAM Dataset notebook

This notebook demonstrates how to fine-tune language models for function
calling capabilities using the xLAM dataset from Salesforce and QLoRA
technique.

Key features:
- Universal model support (Llama, Qwen, Mistral, Gemma, Phi, etc.)
- Memory-efficient QLoRA training on consumer GPUs (16-24GB)
- Automatic model configuration and token detection
- Production-ready code with comprehensive documentation
- Complete pipeline from training to deployment on Hugging Face Hub

✅ Contribution task completed

Author: behroozazarkhalili

notebooks/en/_toctree.yml

@@ -84,6 +84,8 @@
           title: Documentation Chatbot with Meta Synthetic Data Kit
         - local: optuna_hpo_with_transformers
           title: Hyperparameter Optimization with Optuna and Transformers
+        - local: function_calling_fine_tuning_llms_on_xlam
+          title: Function Calling Fine-tuning LLMs on xLAM Dataset