The list of things to be completed before we do the Ewl_Tree2 -> Ewl_Tree conversion
Redraw efficiency sucks
Need to define an mvc_dirty api to allow setting rows/ranges of rows as dirty
ewl_mvc_dirty_index_set(Ewl_MVC *mvc, unsigned int row);
ewl_mvc_dirty_range_set(Ewl_MVC *mvc, unsigned int start_row, unsigned int end_row);
Need to look at the ewl_tree2_column_count_set() call and see if there is a better way to do it.
At this point we have already coded the row/column pairs into the model, so I think it makes sense to split the data count callback into row and column count callbacks.
Need to base the expansion row colours off of the parent colour. This also involves updating the rows below the expansions parent with new colours (and possibly their expansion nodes) if needed.
Currently tree2 stores information on what nodes are expanded in a linked list that is stored in a hash key'd on the expansion points data value. This doesn't work if the data value is NULL. Need to find a new way to key this value.
Horizontal scrolling doesn't work. Always ends up scrolling the first column.
In tree2 test when switching from scrolled to unscrolled headers causing sizing bugs that get worked out on configure.