Tie direction for single-stemmed chords

I've taken a look. There are about as many tie-direction standards out there as there are accidental-stacking standards, stem direction standards, or beaming standards. For tie direction, we've fixed the things virtually everything agrees upon: single voice chord have outer ties in opposite directions, notes in multi-voice context have ties on stem side. After that it gets fuzzier.

Looking at the code, I don't really understand our algorithm, but on appearances, it seems to use the middle line as the determinant, with some additional special handling for seconds. Gould mentions using the middle line something done in "some editions" but recommends against it for reasons she does't explain. This is the sort of thing I guess I could see being a style option some day, just as we have settings to control default beaming, but again, it's easy and 100% to flip stems manually if your preferences happen to differ.

I think more research would be needed before I'd think it worthwhile to change the defaults, but FWIW, it's calculated in one place in the code, just a few lines, so it's probably not not difficult.