Correct some ti-ascii values

[tryashtar]

I wrote a script to compare the ti-ascii values of all tokens with their appearance on a calculator, and found the following discrepancies:

• Y2 - Y5 were appearing as Y3 - Y6
• uwAxes was appearing as SinReg
• Shade𝙵( was appearing as Shade𝙵(α9
• ` was appearing as a slightly different apostrophe character

This PR fixes these

[kg583]

What was your method to find these? A separate font sheet has been on the backlog for years at this point (mostly my fault), which would make verifying on-calc display accuracy a breeze, but in the interim your method could be good for CI.

[tryashtar]

1. I downloaded the images from this page
2. I wrote this script to load the images and produce a hex font file based on unifont's hex format
3. I wrote this script to generate an 8XP program with all tokens and simultaneously generate an image of the tokens using the ti-ascii values and the font definition
4. I loaded the program into jstified, scrolled through the source code, and manually compared each line to its matching line in the image

[kg583]

Oh, that hex font file is exactly the sort of thing I've been preparing (slowly) for the font sheet (from a slightly more cumbersome set of images, as we need to get different languages as well). Didn't know there was a named hex format, though it looks like it doesn't extend easily to color calcs on account of variable-width characters (that also aren't a multiple of 8).

We would appreciate your permission to use the monochrome font file you made as a basis or cross-check for the font sheet, since we do need that font as well. And the PR looks good, so anyone else is free to merge; dunno how uwAxes and Shade𝙵( ended up like that.

[tryashtar]

I don't know anything about other languages, but you're free to use the hex file I generated if it's helpful. I made one for the small graph screen font too. I would have done one for the color calculators too if I could find a similar set of images.

[kg583]

Cool. We'd include the small font as well. I'm not aware of any images like those from TIBD's for the CE; I've been using full screenshots of a program written to display them all.

A few languages replace a few chars in the accented letters range to more appropriate ones for their script; very little total change but it does strictly require a format that distinguishes languages.