Improve node indices

[simonlindholm]

Currently in the randomization code we track node locations by traversing the AST in preorder, and giving every node an index. We should extend that to record every node's start index, end index, and for loops, hypothetical before and after indices used to mark stores/loads in previous/upcoming loop iterations. The extension to loops would help with e.g. the variable expander rewriting for (i = 0; i < n; i++) into for (i; 0 < n; i++).