Graphannis

Changed

- Node IDs in matches don't have the `salt:/` prefix anymore

Added

- Add non-tokenized primary text segments as special labels "tok-whitespace-before" and "tok-whitespace-after" to the existing token
when importing from relANNIS. This allows to re-construct the original relANNIS primary text by iterating over all token in order
and be prepending or append these labels to the token values.
- Add a REST based web-service replacing the legacy annis-service

Fixed

- Load all components when extracting a subgraph using an AQL query

Addded

- Web Service with REST API for the corpus storage
- Copy and link files from the ExtData folder when importing relANNIS.
- Map `resolver_vis_map.annis`, `example_queries.annis` and `corpus.properties` from relANNIS files
to a new unified corpus configuration stored as [TOML]() file. This corpus configuration
is also exported to GraphML.
- Export and import ZIP files containing multiple corpora.

Removed

- Removed Brotli support: use the ZIP file export instead

Changed

- Updated graphANNIS version to 0.26.0

- Support multiple corpora in find, count, count\_extra and frequency

Fixed

- Inverted sort order did not reverse the corpus name list for multiple corpora
- Workaround for docs.rs problems seem to have caused other problems and graphANNIS was not recognized as library

Changed

- Updated graphANNIS version to 0.24.0