Improvement on how to run chronoclust

Previously, config and data files have to be specified via xml file. No XML file is required from now. Function to run ChronoClust now must be directly supplied with the filenames and the configs as parameters.

Make Chronoclust better overall

1. Improve code structure
2. Added feature to sort cluster points before exporting and option to not normalise data
3. Added tests
4. Used numba to improve mathematical calculation

First stable release

The code in this release are the ones that are used to produce all the results in the chronoclust paper.