Open-cravat

- Filter vcf files on Genotype fields and variant call information such as zygosity, VAF, and read depth. Filters can include complex boolean logic, and can be used with annotation data.
- Directly annotate large, many-sample VCFs using `oc vcfanno`. A new vcf is created with `OC_XYZ` fields in the INFO column. Efficiently uses many threads to process genome-scale input.
- Python 3.12 compatibility
- View errors affecting specific variants from the jobs table in the GUI
- Fixes to checking for and installing annotator updates from the GUI
- Improved config file handling for converters

- Refactor submit page javascript to use modules
- Add end-to-end testing
- Improve example input files: add more variants, GUI examples contain same variants and cmdline example
- Exceptions in annotator setup no longer prevent other annotators from running
- Jobs in GUI can still be opened after changing path of jobs directory
- Change settings icon to gear from hamburger
- Add link to docs in header (? icon)

- Fix to summary widgets. They will no longer use the last run filter when reloading the job without a filter selected.
- On the submit page, the annotate button is not fixed to the bottom of the page.
- Restrict aiohttp version to <4.0.0 in anticipation of changes to the aiohttp api

- Add annotation for Exon Number and End Position in mapper
- Fix bug that occurred when using `oc module install` for a module that was already installed

- Fix job database locking issues when quickly refreshing jobs
- Set default temporary file encoding to utf-8
- Put all job databases in WAL journal mode
- Further improve handling of modules with dependencies. Resolve circular dependencies and improve version resolution.
- Allow lowercase ref bases in input files. Cast bases to uppercase in results.
- Add support for filtering on user created notes.
- Remove filtering for certain vcf INFO columns

- Add annotation packages. Packages are bundled sets of annotators and filters designed to support common types on analysis
- Add metrics gathering. Anonymous metrics are collected and reported to KarchinLab. Metrics include information such as input format type, number of variants, annotators selected, and operating system. No indentifiable information is collected about the user, and no details of variants are reported.
- Inform users when installing large (>5 GB) modules through the GUI. Don't install modules if they are bigger than available disk space.
- Improve installation of modules with dependencies. Dependency chains longer than 2 are supported.
- Fix issues when running annotators which have dependencies
- Remove temp files by default in runs submitted through the GUI