Bids-validator

- Fix regression of regex to match recording entity in file names.
- Add a flag to skip checking that any given file for one subject is present for all others. Can help avoid stack limit.
- Have json schemas validate against uri format where applicable.
- Have json schemas only accept *Coordsystem from list of restricted keywords.
- Have json schema conditionally require Description.

- v1.5.5 skipped due to issue with publishing tool.
- Allow datasets to specify HED versions or individual files as validation schema
enhancement
- Fix type column checking in *_channels.tsv files.
- Add EthicsApprovals to dataset_description schema.
- Validate optional fractional seconds in AcquisitionTime in scans.tsv
- Add _recording matches to continuous measurement extensions where applicable in file_level_rules.json.
- Introduce common schema for shared JSON fields, make schemas stricter.
enhancement
- Implement HED value-taking strings.
- Simplify dataset_description schema, enhance phsyio schema.
- Fix cross-talk and fine-calibration extensions.
- Allow top level sbref, json for dwi modality.
- json schema: PowerLineFrequency must be number > 0
- add participants.json to valid_dataset testing data
- Cast dataset desc author values as string when testing for commas.
- Update top level file regex to allow modality metadata files with no required entities.
- Update how we generate filenames of potential sidecars.
- Extend validator for crosstalk+finecalib for FIF
- Update link to contributors
- Allow string n/a for PowerLineFrequency in EEG,MEG,iEEG
- Update contact in Code of Conduct to franklin-feingold
- Make beh and func file level rules consistent with spec.
- Check continuous recording datafiles have matching json.
- Allow optional UTC specifier Z in aqc_time.
- Bump elliptic from 6.5.0 to 6.5.3
- Bump handlebars from 4.1.2 to 4.7.6
- Bump codecov from 3.6.5 to 3.7.1
- Bump lodash from 4.17.14 to 4.17.19

- Validate HED based on all present events files instead of generating list from only nifti files.
- Fix null pointer error in HED sidecar loading code
- Catch case where headers are empty when checking for duration header in tsv files
- Add DataType to dataset_description.json schema
- Fix error with repetitionTime being cast as a number.
- Fixing scans check to make sure BTI and CTF data work (directory based files for MEG)
- Add check for order of entities in filename.
- Allow phsysio and stim tsvs and jsons in *eg directories fixes 990
- Disable old tsv unit checks.
- Add error message if scans.tsv filenames are not all present in the dataset.
- Fix symlink test on Windows
- Use multiplatform compatible regular expressions
- Allow LICENSE file at the root of the dataset
- _part entity renamed to _split following changes to meg spec
- Corrections in the IEEG validation for the MEF3 file format
- Corrected ieeg filetype extension from .mef to .mefd
- [PYTHON] add cross platform testing and real test case
- [PYTHON] add cross platform testing and real test case
- [PYTHON] -> get_path_values method and path.json rules BROKEN + UNUSED

- Improve UTF-8 support in browser environments.
- Upgrade to node-sass 4.13.1 to support node 13 environments better.

- Fix TSV checks returning arrays for evidence values.
- DOC: add TOC + docker info.

- Support large JSON output from CLI without truncation