Enrich2

This version fixes an error that caused combined standard errors to be overestimated for most experiments. Existing users with replicate data should update and re-run their analyses. See the documentation for details on how to do this with least impact.

Major changes:

* Fix standard error calculation for combined replicates.

Minor changes:

* Update manuscript link in documentation.

First post-publication update.

Minor changes:
* Updated Enrich2 citation to point to the Genome Biology paper instead of bioRxiv.
* Updated dependencies to specify pandas 0.18 or 0.19 due to incompatibility with 0.20.
* Added a Conda environment file to simplify setup for Anaconda users.
* Version number is now displayed in the GUI main window title.
* When running on MacOS, the program now sets the matplotlib backend to `'TkAgg'` to avoid the NSInvalidArgumentException that occurs when running in GUI mode with no backend set. Advanced users who want to use a different matplotlib backend under MacOS in command line mode will need to edit the `matplotlib.use()` statement in `main.py`.

New features:
* Added count mode. This allows the user to specify a file containing counts for each element instead of a FASTQ file.
* Added identifier-only sequencing libraries.
* Users can now change the amino acid groupings used for sequence-function maps and diversity maps.

Minor changes:
* Basic SeqLibs now have read trimming options.
* Added a --version command line option.
* Identifier scoring now supports wild type normalization if the special wild type variant "_wt" is present.
* Multiple plotting updates to better support plotting in notebooks.
* Default amino acid groupings for sequence-function maps updated.
* Improved variant validation.
* Fixed an error when maximum N's was set to 0.
* Numerous linting and code cleanup changes.

The version number is now correct throughout the code and documentation.

Initial public release. Corresponds to the version used in the Enrich2 preprint.