feat: add ndpa annotation support

[Mr-Milk]

Add .ndpa annotation file support

Closes #45

Type of changes

• ✨ New Feature (changes that introduce new functionality)

[Copilot]

Pull request overview

Adds support for Hamamatsu .ndpa XML annotation files by parsing NDPA viewstates and converting their nanometer-based coordinates into level-0 pixel geometries so they can be loaded through the existing Lazyslide annotation pipeline.

Changes:

• Implement read_ndpa() (NDPA XML → GeoDataFrame) with coordinate transform based on slide metadata.
• Update load_annotations() to dispatch .ndpa paths to the NDPA reader and to retain non-area geometries (e.g., pins/lines) when min_area is set.
• Add NDPA-focused tests plus NDPA fixture files, and adjust .gitignore to keep those fixtures tracked.

Reviewed changes

Copilot reviewed 5 out of 6 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
src/lazyslide/io/_annotation.py	Adds NDPA parsing/coordinate transform and updates load_annotations() behavior for .ndpa and non-area geometries.
src/lazyslide/io/__init__.py	Exposes read_ndpa as part of the public lazyslide.io API.
tests/test_io_ndpa.py	Adds unit tests covering NDPA parsing, dispatch, metadata requirements, and malformed XML handling.
tests/data/annotations_mixed.ndpa	Adds a mixed-geometry NDPA fixture for tests.
tests/data/annotations_malformed.ndpa	Adds a malformed NDPA fixture for parse-error testing.
.gitignore	Ensures tests/data/*.ndpa fixtures aren’t ignored by broad data/ ignores.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.