Build document-to-study-unit ingestion pipeline

[sonwr]

Summary

Create the first ingestion pipeline that turns uploaded material into study units.

Problem

The product cannot function until raw documents become structured study content.

Proposed solution

Implement a pipeline for parsing source material and splitting it into study units with citations.

Why this matters

This is the foundation for summaries, quizzes, and review scheduling.

Scope

• small
• medium
• large