Glmnet

Fix
* [65](https://github.com/civisanalytics/python-glmnet/pull/65)
Remove `six` dependency entirely.

Changed
* [57](https://github.com/civisanalytics/python-glmnet/pull/57)
Mark the Fortran code as linguist-vendored so that GitHub classifies
this project as Python (37).
* [62](https://github.com/civisanalytics/python-glmnet/pull/62)
Update the cross-validation for users to be able to define groups of
observations, which is equivalent with *foldid* of *cvglmnet* in *R*.
* [64](https://github.com/civisanalytics/python-glmnet/pull/64)
- Python version support: Add v3.8, and drop v3.4 + v3.5.
- Maintenance: Drop versioneer; update and pin dependencies for development.

Fixed
* [55](https://github.com/civisanalytics/python-glmnet/pull/55)
Include all Fortran source code in source tarball; exclude autogenerated C.

Added
* [29](https://github.com/civisanalytics/python-glmnet/pull/29)
Provide understandable error messages for more glmnet solver errors.
* [31](https://github.com/civisanalytics/python-glmnet/pull/31)
Expose `max_features` parameter in `ElasticNet` and `LogitNet`.
* [34](https://github.com/civisanalytics/python-glmnet/pull/34)
Use sample weights in `LogitNet`.
* [41](https://github.com/civisanalytics/python-glmnet/pull/41)
Add `lower_limits` and `upper_limits` parameters to `ElasticNet`
and `LogitNet`, allowing users to restrict the range of fitted coefficients.

Changed
* [44](https://github.com/civisanalytics/python-glmnet/pull/44)
Change CircleCI configuration file from v1 to v2, switch to pytest,
and test in Python versions 3.4 - 3.7.
* [36](https://github.com/civisanalytics/python-glmnet/pull/36)
Convert README to .rst format for better display on PyPI (35).
* [54](https://github.com/civisanalytics/python-glmnet/pull/54)
Use `setuptools` in `setup.py` and update author in metadata.

Fixed
* [24](https://github.com/civisanalytics/python-glmnet/pull/24)
Use shuffled splits (controlled by input seed) for cross validation (23).
* [47](https://github.com/civisanalytics/python-glmnet/pull/47)
Remove inappropriate `__init__.py` from the root path (46).
* [51](https://github.com/civisanalytics/python-glmnet/pull/51)
Satisfy scikit-learn estimator checks. Includes:
Allow one-sample predictions; allow list inputs for sample weights;
Ensure scikit-learn Estimator compatibility.
* [53](https://github.com/civisanalytics/python-glmnet/pull/53)
Return correct dimensions for 1-row predictions, with or without lambda
path, in both `LogitNet` and `ElasticNet` (52, 30, 25).

API Changes
* [10](https://github.com/civisanalytics/python-glmnet/pull/10) the parameter `n_folds` in the constructors of `LogitNet` and `ElasticNet` has been changed to `n_splits` for consistency with Scikit-Learn.

Added
* [6](https://github.com/civisanalytics/python-glmnet/pull/6) expose relative penalty

Changed
* [10](https://github.com/civisanalytics/python-glmnet/pull/10) update Scikit-Learn to 0.18

Fixed
* [3](https://github.com/civisanalytics/python-glmnet/pull/3) ensure license and readme are included in sdist
* [8](https://github.com/civisanalytics/python-glmnet/pull/8) fix readme encoding
* [14](https://github.com/civisanalytics/python-glmnet/pull/14) fix reference to `lambda_best_` in docs
* [16](https://github.com/civisanalytics/python-glmnet/pull/16) fix import path for UndefinedMetricWarning

Added
- Initial release