Make multi-img generation more robust

[d2fong]

Most sbgn glyphs use an array of svg images to render all of the aux. items they have.

Currently, each glyph implementation has a hardcoded set of images and values associated with images, and we only render up to 1 state var and 1 unit of information per node. Ideally we would like to render a few more per node.

There is also more duplication than necessary.

• reduce duplicated code / refactor code to be more robust