CplxTrans (and VCplxTrans?) on negative floats/ints

[sebastian-goeldi]

Hello Matthias,

I don't know how much work it would be, but would it be possible to add or change the CplxTrans, and I think the VCplxTrans would also be necessary, to allow the usage on signed int instead of only unsigned int?

Reasoning being I added this gdsfactory/kfactory#404 and didn't read the API properly before :P (the function there just takes any of the __mul__ overloaded types and returns CplxTrans(layout.dbu) * input (or inverted in the to_um case))

I can work around it, but it would be nicer if the transformations directly support it. Of course only if the change is reasonable and wouldn't jeopardize other API.