[axes proposals] Add new width and weight axes

[davelab6]

@sberlow and I will work on drafting these

[davelab6]

https://docs.google.com/spreadsheets/d/1lZcLW7xo39zG3TTv6EK3iLs5t8OjWWPJv0lb9qEnsDE/edit#gid=2056122370&range=2:5

[dberlow]

Nice. "...constructed by blending other primary axes. "

I'd like to write: "...may be constructed by blending or via referenced instances of other axes."

thinking some more, I edited the value ranges: 0.pops' fine.

1. path, there is a valid difference between these new axes and the original parametric axes, i.e. no negative numbers; I explain this by pointing out that the glyphs of primary parametric axes may have overlapped, or negative values for opaque or transparent features, but one cannot feed negative values through to either the registered wdth, or our new pwdt axes. It makes no sense to users at that point. So, something that is negative in a paramtric axes goes to "0" in pwdt. By then, i.e. once a Width (parametric) hs been made, this only effects the transparency within the pwth axes, and that could be 0, thus the edit. let me know.
2. I rewrote and revalued grad, as a relative local +/-? let me know.
3. I filled in the "blend participants" but thisynoclearo yet because in one column, we are combining "what's use to make me" and "what I'm used to make"

[dberlow]

1. path, there is a valid difference between these new axes and the original parametric axes, i.e."

spell checker...

1. pwth, there is a valid difference between these new axes and the original parametric axes, i.e.