* update seqspec check

* add spec parameter back to check function

* support gzipped yaml file for function load_spec

* fix bug in function run_check

* support gzipped yaml file for function load_spec

* ignore onlist in seqspec check when needed

* fix bug in seqspec check

* code review

* Update tox to use newer python interpreter versions

* validate_check_args now returns a list

* Deal with ascii and png display function name changes

* protocols and kits are a controlled vocabulary now.

* Clear out the environment variables before running the tests

In case they happen to be set

* Rename files from .txt to .tsv ot better match DACC conventions

* Set a specific seqpec version as the structure keeps changing

* Structure now needs an files attribute

* Update test for remote access

needs to change more text in the example, and suppress the new call to
the network

* Update for more detailed onlist structure

many calls to create Onlists needed more attributes

* update onlist test to use preferred -i argument

-r was deprecated

* Reduce code repetition by using to_dict in __repr__

The return the dictionaries in the same order, might as well have
fewer places to update

* Be robust to missing values for the File and Read objects

All of the attributes for the File object are retrieved with getattr,
and the Read.files attribute introduced with 0.3 is protected with
getattr.

You do need to provide a default value with getattr if you want to
avoid an attribute error

* Add files object to example seqspec

With constrained_layout and a single modality the height of the bar
graph showing the regions collapsed into almost a line.

This plots it without constrained_layout, and adjusts the title offet
as needed.

This fixes #44

…x to force single end reads for read with max size

… dev-requirements.txt, simplified release process in the Makefile, simplified version tracking with pyproject.toml through setuptools_scm

…t and MANIFEST.in

…ic Base Class. this removes the need to specify yaml tags. These now get stripped. Changed formatter from black and flake8 to ruff.

…it in spec, it wasn't being used anywhere, removed seqspec convert from the cli (currently not implemented)

- updated seqspec index to initialize pydantic models with named args
- removed tox, changed to pytest as a test manager
- cleaned up internal api for seqspec onlist (todo, add subcommands list, download, join)

…dated version (so subsequent loads of the file work). TODO consider using loose loading only for format and check commands, and strict loading for every other command.

…loading validation errors and print to stdout when trying to load with strict mode

…map primer id from a read to any level in the library spec. no changes made to seqspec index, but greatly expands style of specs that are compatible.

… use updated functions, updated seqspec index to fix file name useage when -s file is specified, updated some tests that were previously incorrect in seqspec index which used read ids instead of region ids. seqspec index was changed to fix the behavior when asking for region indices

…nlist/find, updated seqspec index to use consistent internal types, expanded tests for index

…ates to index no longer require primer_id to be in leaves

…d within each read are fully unique

* added sgrna_target as region_type

* fixed format_kallisto_bus_force_single

…pec_fn from seqspec check, made seqspec insert and seqspec modify consistent (taking in list of *Input objects), annotated *Input objects for llm usage

…subregion_sequence which checks that the min/max length of a region are equal to the sum of the sub regions and that the sequence of the region is equal to the concatenation of the subregions. suggestions by Zhewei Shen and Ian Whaling based on a spec submitted by Alex Barrera to IGVF portal (https://data.igvf.org/configuration-files/IGVFFI9197UDXC/)

…ot overwrite sequences if present in region

…s don't exceed sequenceable range given by the library elements after or before the primer id (based on the strandedness of the read)

…spec