Add support for textual content of canvases

[tomcrane]

It would be great if, once you get to the level of the canvas, you can continue using the API to extract and render the textual content.

There's a very crude idea of what this might look like here:
http://tomcrane.github.io/wellcome-today/annodump.html?manifest=https://wellcomelibrary.org/iiif/b28047345/manifest

(tick the "lines of text" box and navigate through the work to some printed content).

Having helpers like getTextLines(..), getFullText(..) and so on would help application developers pull out the text content for reuse.