Request: Add LLM fine-tuning section (LoRA, QLoRA, RLHF)

[aasimansari1]

Would love to see a section covering:

• LoRA (Low-Rank Adaptation)
• QLoRA (Quantized LoRA)
• RLHF (Reinforcement Learning from Human Feedback)
• DPO (Direct Preference Optimization)
• Common interview questions around these techniques

These are becoming increasingly common in ML interviews at top companies. Great repo btw, definitely one of the most comprehensive I've seen!