Shakenbreak

----------
- Add `snb-mag` function, and automatically check the magnetisation from `ISPIN = 2` `OUTCAR` files when continuing
relaxations with `snb-run` (and change to `ISPIN = 1` if magnetisation is negligible).
- Update handling of minimum distances and oxidation states, to deal with single-atom primitive unit cells and
systems where `pymatgen` cannot guess the oxidation state (e.g. single-elements, intermetallics etc).
- Docs updates

----------
- Make parsing of `DefectEntry`s more robust.
- Update dependencies (now supporting `python=3.10` due to `numba` updates)
- Refactor `CITATION.cff` to `CITATIONS.md`
- Update docs, formatting and cleanup.

----------
- Update `numpy` requirement to `numpy>=1.21.2` to fix `numpy.typing.NDArray` import error.
- Add News & Views free-to-read link to docs

----------
- Updates to `snb-run` (copy `job` from parent directory if present, switch to `ALGO = All` if poor electronic convergence...)
- Make `format_defect_name()` more robust
- Update docs and `README.md` with published article links
- Formatting and cleanup
- Make oxidation state guessing more efficient (previously was causing bottleneck with large cells)
- Fix oxidation state guessing for rare elements
- Add note to `Tips` docs page about bulk phase transformation behaviour
- Refactor to `json` rather than `pickle`

----------
- Change `numpy` version requirement in `docs/requirements.txt` to `numpy>=1.21` to work with `numpy.typing.NDArray`.

----------
- Refactor Distortions() class to take in DefectEntry objects as input, rather than Defect objects, to be
compatible with `pymatgen-analysis-defects`.
- Fix ticks and ticklabels in plots