Copasul

* `Copasul()` class added (please see `example_call.py` how to integrate it into python code)
* added: `copa["export"][myFeatSet] = pd.DataFrame()` so that feature output is stored in generated copa dict. myFeatSet: `glob, loc, bnd, voice, gnl_f0, gnl_en, gnl_f0_file, gnl_en_file, rhy_f0, rhy_en, rhy_f0_file, rhy_en_file`
* output table columns (and dataframe columns in `copa["export"]`) are now alphanumerically sorted
* dict `config.plot.browse.grp`, which is processed in `copasul_plot.plot_browse()`. This dict contains zero or more `myGroupingKey-myGroupingValue` pairs. Each `myGroupingKey` should match one of the strings in `config.fsys.grp.lab`. By this the user can select to plot images with (a combination of) certain grouping values only.

* augmentation log messages redirected to log file

* f0 contour is now horizontally extrapolated till 0.00 and not 0.01 By this summary statistics of the `glob` and `gnl` feature not contain the same values.
* `copasul_styl.styl_polyfit()`: polyfit made more robust

* filter variables `is_init(_chunk)` and `is_fin(_chunk)` now treated as factors in R output

* new voice quality features
* **jitter:** `m, sd, m_nrm, sd_nrm`: dur mean and std (+ normalized over file); `jit_nrm`: normalized over file
* **shimmer:** `m, sd, m_nrm, sd_nrm`: amplitude mean and std (+ normalized over file); `shim_nrm`: normalized over file

* default config file removed; now hard-coded in `copasul_init.py`