Riassigner

Changed
- updated _assign_ri_value function to delete the RI key if the value is not greater than 0 [122](https://github.com/RECETOX/RIAssigner/pull/122)

Added
- reading RI values from the comments field [112](https://github.com/RECETOX/RIAssigner/pull/109)
- column name sanitizing and mapping using matchms [118](https://github.com/RECETOX/RIAssigner/pull/118)
- support for parquet and mgf formats [116](https://github.com/RECETOX/RIAssigner/pull/116)
- functions to check whether RI and RT are present [115](https://github.com/RECETOX/RIAssigner/pull/115)
Changed
- updated dependencies to newest versions [116](https://github.com/RECETOX/RIAssigner/pull/116)
- changed build system to use poetry [109](https://github.com/RECETOX/RIAssigner/pull/109)
Removed

Changed
- pinned pint >=0.17,<0.20 and matchms >=0.14.0,<0.18.0 for dependency issues

Added
Changed
- replaced asserts with `ValueError` as asserts can be disabled. [91](https://github.com/RECETOX/RIAssigner/pull/91)
- updated matchms dependency to `>= 0.14.0`. [90](https://github.com/RECETOX/RIAssigner/pull/90)
Removed
- removed test cases which include data that has `None` retention times. [90](https://github.com/RECETOX/RIAssigner/pull/90)

Added
- data/Data.py: Added `update_possible_rt_keys()` and `update_possible_ri_keys()` methods for `Data` class.
Now users can supply their own identifiers for RT/RI columns (csv files) and identifiers (msp) in their data. [74](https://github.com/RECETOX/RIAssigner/pull/74)
- General: Added missing documentation and updated [README](README.md). [80](https://github.com/RECETOX/RIAssigner/pull/80)
- setup.py: Added versions for dependencies. [80](https://github.com/RECETOX/RIAssigner/pull/80)
- utils.py: Added method to check if array is sorted. [82](https://github.com/RECETOX/RIAssigner/pull/82)
- data/SimpleData.py: Added class for in-memory data which is read-only. [82](https://github.com/RECETOX/RIAssigner/pull/82)
Changed
- data/Data.py: Unified default RT/RI column identifiers between `PandasData` and `MatchMSData` classes. [74](https://github.com/RECETOX/RIAssigner/pull/74)
- data/MatchMSData.py: `MatchMSData` class now looks up for RT and RI identifiers from within default identifiers list
instead of hardcoded 'retentiontime' & 'retentionindex' keys. [74](https://github.com/RECETOX/RIAssigner/pull/74)
- compute/Kovats.py: Computation is now real piece-wise linear interpolation and doesn't assume a step width of 100 but interpolates in the actual interval. [76](https://github.com/RECETOX/RIAssigner/pull/76)
Removed
- data/Data.py: Removed `read` method from class and contructor -> all child classes must now explicitly implement and call the superclass constructor. [82](https://github.com/RECETOX/RIAssigner/pull/82)

Changed
- data/PandasData.py: `_read_into_dataframe()` now deduces a proper delimiter via Python's [`csv.Sniffer`](https://docs.python.org/3/library/csv.html#csv.Sniffer) class. [73](https://github.com/RECETOX/RIAssigner/pull/73)