Correct encoding usage

[danshearer]

Access uses UCS2, which pre-dated UTF-16, and we should take care in conversions in Access-Tool. We don't expect there to be two-character glyphs in Spyderisk Access source code, but Access might still notice a corner case.

Similarly, we should take care to use UTF-8 not ASCII, even though they may be the same thing most of the time. We are just one non-standard quote mark or arrow away from needing UTF-8, so we should be using that.