Typed rects with negative sizes should return as an empty rect

[changm]

This change is 

[kvark]

It's a feature that is small but needs a somewhat broader consensus (RFC?).
Are there any users that rely on negative rectangles?

[mrobinson]

It seems that if #177 is implemented this won't be an issue, because negative lengths won't be possible.

[kvark]

@mrobinson but we'd still have negative rectangles

[changm]

So do we try for an RFC?

@kvark I have some issues in WR dealing with empty rects right now. I can work around them if that's ok and clean them up once this is fixed?

[kvark]

@changm I don't feel myself in a position to make this decision for euclid, so unless someone steps in and approve it, I suggest working around it without modifying euclid.

[nical]

There hasn't been activity for a while, so closing this for now. We can open a new PR/issue if we still want to use check for negative rects, although doing so would require carefully checking WebRender/Servo's code to make sure it doesn't introduce subtle breakage.