Optimise external secondary instance memory use

[seadowg]

As we did for entities with #5623, we should optimise all external secondary instances so that they do not need to be loaded into memory for simple eq filters (child = 'value'). There's some follow on work that we do as a consequence of this:

• JavaRosa should no longer not attach instances if they aren't referenced by an instance expression as pulldata and an equivalent instance expression will have the same memory usage anyway.
• As part of this, we should also remove the specific database backed implementation of pulldata and reimplement the function so that it's directly sugar for XPath (like in this gist).

Questions

• Should this be multiple issues?
• How do we deal with search? Should we reimplement it as sugar like with pulldata or do we just remove it?