Falwa

- Remove the use of `setuptools` in `setup.py`
- Dependencies in `environment.yml` matches MDTF
- Dependencies in `setup.py` changed (for the sake of MDTF incorporation)
- Deployment onto Conda channel (at least for Linux) is included in `notes/developer/conda_deployment.md`
- Deployment using pip (only include source code)
- Migrate completely to python 3.10 (TODO: fix GitHub workflow)
- Match MDTF package version requirements in pip
- Deployed on [https://pypi.org/project/hn2016-falwa/0.7.1/](https://pypi.org/project/hn2016-falwa/0.7.1/)

- Now `northern_hemisphere_results_only` becomes an input argument for `QGField`. Please initialize this when creating `QGField` object and remove this argument from the methods (`compute_reference_states` and `compute_lwa_and_barotropic_fluxes`).
- `eq_boundary_index` is now an input arguemnt merely for `QGFieldNHN22`, but not `QGFieldNH18`. Remove it if you are using `QGField`/`QGFieldNH18`. You can refer to the jupyter notebook examples for the updated usage.

This minor release include a fix in graph plot procedure `scripts/nhn_grl2022/graph_plot_module.py` - the plotting script for the residual plot is fixed such that it produces the same plot on different machines.

This is a minor release that fixes the issues with `setup.py` that cannot locate the compiled fortran modules. Thanks chpolste for fixing this!

The archive on PyPI has also been also fixed in v0.6.5.

This release includes the fix for the calculation of meridional flux component. See the [correction note](https://github.com/csyhuang/hn2016_falwa/blob/aa45499fcfb0bdc0236f93b39c83bd9f68ef9abb/notes/LWA-budget-on-the-sphere-revised-20221230.pdf) for details. The corrigendum on Neal et al 2022, GRL corresponds to this release.

The main change in the calculation can be found in this commit: 38d95f6202f2b0d052263437a61124fe3ace0d65

Please remove all the compiled fortran modules on your machine and `git pull` the newest version of fortran codes for installation (`python setup.py develop`).

There were resolution-dependent procedures which worked with the (360x181) grid we used for NHN2022, GRL. This bugfix enables the new inversion routine to take data of other grid sizes.