Feature request: aggregate of non-ASCII/system file/directory characters

[kieranjol]

I'm enjoying digging into the lists of non-ASCII and troublesome characters. I think it could be useful to show an aggregate of the characters that appear in a report, and how often they appear. For example:

 characters outside of ASCII range: '0xc9, LATIN CAPITAL LETTER E WITH ACUTE: É' (248)
 characters outside of ASCII range: '0xf028, None: ' (4)
 non-recommended character: '0x5b, LEFT SQUARE BRACKET: [' (1474)

It could be useful so that during appraisal, if I know that latin letters with acutes are supported within the repository, but perhaps bullets or other characters are not, then it would speed up the process of identifying problematic characters.

[ross-spencer]

Sorry it took a while to get back to this. It's a good suggestion. I need some sample data so finally got round to creating a proper repo for some of my other test work here: https://github.com/ross-spencer/big-list-of-naughty-files which generates a lot of output that will appear in these kinds of aggregates. Unfortunately it breaks a few more things so I'll try and fix those first then add some more sample data to this issue to create the aggregates.