Release v0.5.0

[github-actions]

Bumps the version from 0.4.0 to
0.5.0 in pyproject.toml and docs/source/conf.py.

Changes in this release

• decreasing python version floor to 3.11
• test fix
• fixing check of hash for multiple file dsets
• Revise README
• updating techniques and fixing wordings
• updating wording in example
• updating descriptions of ptycho datasets
• fixing bug of zenodo updates not checking weights checksum changes
• fixing multiple download pbar titles
• making it easier for datasets to contain multiple files (and a single head/main file)
• bugfixes - stale failed download, shared location w/ multiple files
• handling 7z files
• handling zenodo being down or other download failures
• allowing for files to be selected from within .zip on zenodo
• adding to the PR action a note about what items were added to the weights release
• fixing docs header
• fixing versioning
• cleanups
• fixing typechecking errors
• cleaning up docs
• fixing download pbars
• Add dated weights versions from the weekly check
• cleanup
• cleanup
• simplifying metadata
• cleaning up widget
• pruning tests
• simplifications
• cleanups
• simplifying dataset
• simplifying config
• fixing CI
• index guards
• Opus code-review bugfixes
• fixing check of old/removed dsets
• removing a weights
• simplifying new data submission
• Add index entry from issue [New Dataset]: peak detection polymers #12
• switching to issue page only
• making optional upload more explicit
• improving download link parsing
• Removing the docs PR submission method and changing to GH Issue only
• Add index entry from issue [New Dataset]: TutorialUNet_instance #8
• more helpful error messages for issues
• hopefully fixing issue automation
• updating Add Dataset for links
• handling zenodo being down lol
• updating techniques for ML, hopefully fixing docs cache issue
• allowing multiple techniques
• Add dated weights versions from the weekly check
• hopefully fixing build check failing
• fixing formatting of weights updates
• checking for outdated weights on download()
• handling updated weights on zenodo -- not archiving on gh
• fixing bug in weights release
• adding weights versioning
• config: one locations mapping with a reserved personal entry; drop data_dir, stores, set_data_dir
• allowing for passing destinations to download--but need to be simplified
• hopefully fix docs test
• index: write kind: dataset explicitly in every entry and generator
• forms: url, kind, version and model fields in the web and issue forms; fix issue workflow trigger and body quoting
• weights: url field for opaque download links, TutorialUNet_v1 entry, notebook demo
• weights: kind/version/model schema fields, list_weights, docs page, CLI and template support
• config: dask-style loader, named stores, cache-dir default; doc/ -> docs/
• rename: em_database -> emdatabase, datasets/ -> index/, docs builder out of the package
• fixing for gh-pages
• cleanup and docs updating
• adding search functionality in notebook without a widget
• fixing pbar bug, better distinction between shared and user dsets
• updating metadata to dataclass and replacing size with size_bytes
• typing to 3.10+, DownloadFuture -> DatasetPath refactor
• fixing _PENDING not passing to other paths, subclassing issue, adding a location field distinguishing shared vs user files
• linting cleanup
• adding pre commit and updating pyproject.toml
• Release: Bump the version to 0.4.0
• Docs: identical top bar on every page, full width, bigger fonts, simpler
• Feature: redesign the docs site as a widget-style app
• Settings widget: Browse to the right of the field, no "upload" prompt
• Settings widget: client-side folder Browse
• Feature: an editable settings widget
• Docs: document the settings model and shared data
• Feature: matplotlib-style settings + system-wide data resolution
• Release: Bump the version to 0.3.0
• Fix directory tests for the background-download default
• Re-anchor download toasts on every ds.download()
• Fix toast cancel and add toasts for bare downloads
• Feature: Make the docs landing page an interactive dataset browser
• Feature: Non-blocking downloads and an interactive Jupyter widget
• Release: Bump the version to 0.2.0
• Feature: Add a prepare release workflow
• Feature: Add the PdCuSi crystallization dataset
• Fix: Stop CI downloading ~150 MB per matrix leg to test two datasets
• Feature: Add GitHub Actions workflow to publish to PyPI
• Feature: Package metadata for PyPI release
• Feature: Add five Direct Electron datasets and six eXSpy EELS datasets
• Feature: Add GitHub Action workflow and script to convert new dataset issues to YAML
• Fix: Rename ISSUE_TEMPLATE.yaml to new_dataset.yaml and update template formatting
• Fix: Update ISSUE_TEMPLATE.yaml field names and formatting for consistency
• Documentation: Playing around with creating a table of all datasets.
• Remove uv from project dependencies in pyproject.toml
• Fix: Remove debug print statements and correct datasets path in Sphinx configuration
• Fix: Improve error handling and logging during Sphinx build for datasets
• Enhance: Add debug prints for dataset path verification during Sphinx build
• Fix spelling and update datasets link in README
• Add init for packaing yaml files.
• BugFix: Correct path concatenation in test cases for consistent directory structure
• BugFix: Correct path concatenation in test cases for consistent directory structure
• Fix: Update documentation paths in GitHub Actions workflow for consistency
• Refactor: Update paths to use pathlib for improved readability and maintainability
• New Data: Add GitHub Actions workflow for building and testing across multiple Python versions
• New Data: Add GitHub Actions workflow for documentation deployment
• Refactor: Improve CSS styles for better dark/light mode support
• New Data: Update documentation and improve organization.
• Refactor: Refactor and add documentation.
• BugFix: Update stubs and handle progress bar.
• New Data: Added new datasets and update user agent for zenodo.
• Testing: Add tests for downloading and changing the file locations.
• Documentation: Playing around with creating a table of all datasets.
• Add README.md
• First Commit

To publish

1. Merge this PR.
2. Draft a release
   tagged v0.5.0 against main.
3. Publishing it triggers publish.yml, which checks the tag against
   pyproject.toml and uploads to PyPI.

CI does not start on this PR by itself - it was opened by a workflow. Close
and reopen it if you want build.yml to run before merging.

Opened by the prepare release workflow.