Mr-clean

* Re-ordered changelog to have most recent version on top.
* Changed functionality of summarize
* Summarize function was doing too much: moved some of the extra stuff to a new diagnose function
* It now no longer gets df.info() information or datatype information
* The parameter preview_rows now specifies the amount of rows to display in the preview. The format has changed to the following, where *n* represents the minimum of the value specified by preview_rows and the total amount of rows in the DataFrame:
* floor(n/4) header rows
* n - 2 * floor(n/4) rows sampled from the middle of the DataFrame (excludes header and tail rows)
* floor(n/4) tail rows
* Added diagnose function
* Subset of functionality originally in 'summarize' function.
* Preview of dataset, df.info(), df data types, and potential outliers
* Preview of dataset works exactly how preview works in new summarize function
* Added clean function
* Reformats column names to lowercase, with underscores instead of spaces
* Tries to remove whitespace and extra characters when possible
* More features coming soon

* Added documentation for methods.
* Added corr_matrix function that expands on the functionality of pd.DataFrame.corr by adding support for categorical and boolean data.
* Added flexibility to the summarize function.
* It now outputs describe for both numeric and non-numeric data.
* It also now outputs a less noisy version of the outlier DataFrame, while maintaining the same amount of information
* Finally, it now also supports outputting to a folder, and printing the entire DataFrame outputs as .csv files.

Changed package structure again, and added more statistics methods.

Changed the package structure to allow for easier development in future.