Add comprehensive documentation explaining document indexing process

[Copilot]

This PR adds detailed documentation explaining how documents are indexed in the AI Hub solution, addressing the need for clearer explanation of the document processing pipeline.

Changes Made

Enhanced Azure Cognitive Search Concepts Documentation

• Added comprehensive "Document Indexing Process" section to docs/content/en/docs/Concepts/azure-cognitive-search.md
• Documented the complete pipeline with 5 key stages:
  1. Document Upload and Storage - Azure Blob Storage integration with PDF page splitting
  2. Text Extraction and Analysis - Azure Form Recognizer for structure-preserving text extraction
  3. Content Chunking and Sectioning - Intelligent text splitting with overlap for context preservation
  4. Vector Embedding Generation - OpenAI text-embedding-ada-002 model integration with batch processing
  5. Search Index Population - Azure Cognitive Search index creation and population

Detailed Technical Specifications

• Documented search index schema with field descriptions and properties
• Explained search capabilities: vector search (HNSW algorithm), semantic search, and hybrid search
• Added configuration parameters: chunk sizes (1,000 chars max, 100 char overlap), embedding dimensions (1,536), etc.
• Referenced the prepdocs.py implementation for technical details

Enhanced Chat on Your Data Use Case

• Added "How It Works" section to docs/content/en/docs/Use Cases/chat-on-your-data.md
• Provided high-level overview of the indexing process for end users
• Cross-referenced detailed technical documentation for developers

Benefits

• Users now understand how their documents are processed and transformed into searchable content
• Developers have technical details needed for customization and troubleshooting
• Clear explanation of the AI-powered search capabilities and underlying technologies

The documentation builds successfully and renders correctly in the Hugo-generated site.

Fixes #72.