I thought there was a direct method but I cannot find it. That leaves two methods
1. What Acrobat has to do itself. Start with a Page object, and work up the tree using Parent, Kids, Count to identify it.
2. Build a table of page numbers and objects using PDDocGetPageObjByNum
I see at several issues with your code.
(a) You are treating Names as a simple array, but it is actually a name tree. It will only be a simple array in PDFs with very few destinations.
(b) You are skipping other cases, such as the case where the name tree value is the destination array already.
(c) Per 12.3.2.2 (which is not very clear, but I have seen it) the first entry in a destination array can be an integer (which is useful because it is what you are looking for, though I'm not sure if it's zero based).