v0.4.0 - Basic Docs and Misc.

Create initial user guide/tutorial/examples for ReadTheDocs.

Use mkdocstrings to generate the API documentation.

Change the exported names of GadgetishHDF5Dataset and PackedTree to clarify API
Make several methods/functions private since we will be exposing more of the module
tree than originally planned.

Big Breaking Change: KDTree is now OpTree (since it's not actually a KDTree...).

Several changes to the HDF5Dataset, Cubes, and InMemory dataset APIs, see relevant
commits (ffa687e, d26db2a, 4c87748, 800fe80, 4fe6f14, 8bfe1e6, 74f00fa)

[0.4.0] - 2026-04-24

🚀 Features

• Allow setting the initial particle type (ffa687e)
• Add explicit init (2fae868)
• Change default sorted_filepath name (d26db2a)
• Add sorted_filepath and data_slices to public properties (45b3f11)
• Allow specifying initial particle type to load (4c87748)
• Add options to dataset saving (800fe80)
• Allow specifying the sorted filepath on cubes call (4fe6f14)
• Allow specifying only a single particle type to load (8bfe1e6)
• Change default for save_dataset (74f00fa)

🐛 Bug Fixes

• Fix attempting to load cubes from non-hdf5 files (ad280ae)
• Scripts not generated correctly (81b1b44)
• Don't save InMemory datasets by default (66c0268)

🚜 Refactor

• Switch KDTree to OpTree in code (8626893)
• Rename kdtree.py to optree.py in performance tests (9cea29a)
• Rename test_kdtree.py to test_optree.py (76ada83)
• Rename kdtree.py to optree.py (d8659af)
• Add missing typing (4ba4f50)
• Remove json_parsing.py (7695160)
• Make save_fig private (773982f)
• Add _json_parse.py (425c5ea)
• Make internal functions private, simplify name, signature (a1cbf27)
• Make cubes_dict private (259a6ae)
• Add missing typing (56fee08)
• Move several internal functions to private (5740d1c)
• Exposse create_metadata (f7309ec)
• Change multiparticle type Cubes behavior (3ca1c30)
• Remove duplicate parallelism test, only check thread number (b407d60)
• Make json_parsing module private (5143c94)
• Change exported names (3aa1af9)

📚 Documentation

• Clean up docstrings (bac216c)
• Fix reference to MultiCubes (24ac21c)
• Add link to documentation (635c094)
• Add initial Working With Datasets tutorial (5cafbed)
• Improve Getting Started page (9e27523)
• Improve Getting Started page (92038ac)
• Mention ParticleCubes in OpTree docstrings (d9e57a7)
• Remove DataContainer from general Data Objects section (bdf7161)
• Break up monolithic bounding_box import (77c959f)
• Provide numba placeholder information (5e41283)
• Break up monolithic CONTRIBUTING snipped (b181eb6)
• Remove installation instructions from Getting Started (88eb88d)
• Tweaks to Citations.md (91afa6b)
• Manually set navigation in zensical.toml (c445aae)
• Update missing icons/links/pages on indices (141b82b)
• Add missing icons and page titles (9085c2a)
• Add missing doc files (9379448)
• Change layout and section index pages (aff0e0f)
• Add warning about data fragility (8df602a)
• Upload initial versions of several docs pages (2ce961a)
• Update the welcome page (8b0d703)
• Improve numba section (aa75cf7)
• Remove 'The' (70f0670)
• Update README to match Cubes changes, add snippet marks (037bcda)
• Add basic Code and Contributing (3786ad7)
• Reorganizing documentation (fc96ca0)
• Remove spaces from file names (f6ac9f4)
• Improve README readability (83a8a3b)
• Add basic API pages (0ee8286)
• Update zensical config (d13041f)
• Add missing module docstrings, init.py (c4b62ad)
• Convert to Numpy style and add missing documentation (32cb38b)
• Add missing ruff flags in prep for docstrings linting (25b8a0d)
• Downgrade packed format version (654ac2d)

🎨 Styling

• Turn on docstrings checking, switch to numpy style (cf27eb3)
• Minor style fixes from ruff (8817bbe)

⚙️ Miscellaneous Tasks

• Update zensical and add zensical tasks to docs feature (fb3c675)