[RESTRUCTURE] Restructure course for separation

[burtenshaw]

This PR restructures the course by removing some units. The objective is to focus on 'LLMs'. The resulting structure will be:

• Chapter 0: Setup
• Chapter 1: Transformer models
• Chapter 2: Using 🤗 Transformers
• Chapter 3: Fine-tuning a pretrained model
• Chapter 4: The 🤗 Tokenizers library (was Chapter 6)
• Chapter 5: Classical NLP tasks (was Chapter 7)
• Chapter 6: Fine-tune Large Language Models (was Chapter 11)
• Chapter 7: Build Reasoning Models (was Chapter 12)

In separate repose I will migrate these chapters:

• Move unit 5 and 10 to a new ‘datasets course’
• Move chapters 4,8,9 to a new ‘hugging face 101’

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.