Documentation Improvement: Add Descriptions and Context for Features

While using Docling, I noticed that the documentation mostly consists of code snippets without any explanation of what the modules do or how they work. Features like OCR, image extraction, and multimodal pipelines are mentioned, but there’s no description of their purpose, usage, or expected outputs.

Adding short explanations (e.g., what OCR does, when to use certain pipelines, supported formats) would make the documentation far more helpful, especially for new users.

Alternatives
Right now, the only way to understand the system is by reading the source code, which is time-consuming. Improving the docs would make the tool more accessible and easier to adopt.

Review the existing Docling documentation for the OCR, image extraction, and multimodal pipeline sections mentioned in the issue. Identify where explanations are missing, then document each feature's purpose, use cases, supported formats, and expected outputs; done means new users can understand these features without reading the source code.

Written by the indexing model from the issue text.