Pygenclean

This release fixes a small issue in the automatic report (missing value for HH missing at the moment of creation of the report).

This release fixes a small issue in the automatic report (logic error for samples and markers missing rates in the summary result table).

This release fixes an issue with the installation.

Here are the changes for the 1.8.0 release:
- It is now possible to estimate sample contamination using [`bafRegress`](http://genome.sph.umich.edu/wiki/BAFRegress).
- Scree plot was too long in the automatic report; it has been reduced to fit the report page.
- Only autosomes are used for creating _genome_ file (IBS and MDS).
- Better logging capability.
- Updated documentation.
- Correction of minor bugs.

These are the changes in the next release:
- Improved content in the automatic report.
- Improvement 7, where the list of significant markers for the plate bias step is no longer reported, but only the number of significant markers for each plate is.
- Improvement 8, where the samples were sorted in the sex check step so that the order is consistent between the table and the different figures (if any).
- Improvement 9, where a table summarizing the status of every related pairs found is shown in the report.
- Now possible to change the title of the automatic report (using the `--report-title` option).
- Better [documentation](http://lemieuxl.github.io/pyGenClean/).

Here is a description of the bugs that were fixed in the next release:
- Fixes issue 4, where the automatic report failed if there were no duplicated markers in the dataset.
- Fixes issue 5, where the automatic report failed if there were no duplicated samples in the dataset.
- Fixes issue 6, where the scree plot is not included in the automatic report.
- Fixes issue 10, where there was a problem when Unicode characters were present in figure paths.
- Fixes issue 11, where the pipeline failed if ethnicity was checked without reference populations.
- Fixes issue 12, where the wrong number of markers was included in the report. Now, only the number of _unique_ markers is shown.
- Fixes issue 13, where there was a problem with all paths when running the pipeline on Windows operating system.
- Fixes issue 15, where the `pyGenClean_merge_reports` script fails when a directory contains a trailing slash ("`/`").
- Fixes a small bug, where there was an issue in the automatic report if only extract/exclude or keep/remove was used the in `subset` script.
- Fixes a small bug, where output files with no data contained an empty line instead of being empty.
- Fixes a small bug, where the script couldn't find a sample because it was sanitized for LaTeX.

These changes will be included in the 1.7 version of `pyGenClean`:
- automatic report
- produce an automatic report (PDF using LaTeX)
- script to merge multiple reports (for different steps of the QC pipeline)
- summary files for a QC run including:
- list of samples that were removed (with reason)
- list of markers that were removed (with reason)
- ethnicity module
- possible to exclude the three reference panels when computing MDS values
- bug fix when the three reference panels do not have the same list of markers
- updated documentation