Restructure marker and paddleocr packages

[p-j-smith]

Follows on from #42:

• add packages/ocr/marker and packages/ocr/paddleocr packages as uv workspaces
• refactored the pyonb_paddleocr API and Dockerfile to be consistent with the other OCR tools in pyonb
• update pyonb_marker and pyonb_paddleocr Dockerfiles to cache dependencies
• use DATA_FOLDER for mounting the marker and paddleocr data volumes
• update docs on installing pyonb and running inference with marker and paddleocr
• tested locally and confirmed tests still pass for marker and docling

[Copilot]

Pull Request Overview

This PR restructures the marker and paddleocr packages by moving them from the src/ocr/ directory to dedicated workspace packages in packages/ocr/. The change improves dependency management by using uv workspaces and updates Docker configurations for better caching.

• Converted marker and paddleocr from nested modules to standalone workspace packages
• Updated Docker configurations to use multi-stage builds with dependency caching
• Simplified environment variable handling using DATA_FOLDER consistently across services

Reviewed Changes

Copilot reviewed 21 out of 25 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
src/ocr/paddleocr/*	Removed old paddleocr implementation files
src/ocr/marker/*	Removed old marker implementation files
packages/ocr/paddleocr/*	New paddleocr workspace package with pyproject.toml
packages/ocr/marker/*	New marker workspace package with simplified API
src/api/app/routers/*.py	Updated to use aiohttp for async requests and simplified environment handling
docker-compose.yml	Updated build contexts and volume mounting for new package structure
pyproject.toml	Added new workspace members and optional dependencies

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}