Add further checks and error messages when the input dataset isn't in the right format #384
Description
For users who do not check how the data should be input, or for those who struggle or fail to see the dataset examples, add further checks and error messages when the input dataset isn't in the right format:
- If there are any empty data rows (it currently reports error about non-forced distribution, because it obviously finds Q-sorts that don't conform to the predominant set of values; should specify that there are "NA" values and these need removing)
- (at least for GUI) If the first row of data starts in line 3 or more, i.e. the CSV file had 2 or more leading lines with non-data characters (the error message given currently is generic; it'd be more useful if it said something about the need to delete the top rows)
- (at least for GUI) If the first column or row contain statement or Q-sort ids, no error is given but the analysis cannot run. Consider adding a check