The goal in 4.0 is that there'd no longer be any significant benefit to an official over a user-created dictionary, i.e., you won't need our version of those dictionaries because a user-created one will support all of the same formatting / variants / etc and be comparably fast.
Official support is dicey because it'd require quite a bit of extra coding to support those romanization systems in official databases, for something we'd be giving away for free - we were able to justify that with Cantonese because we also had several paid Cantonese dictionaries to offer (and more coming), but unless somebody comes out with a line of paid Minnan / Hakka / etc dictionaries or MoE changes their license to allow us to charge for them it would be tough to make that a priority over the several years' worth of other items on our to-do list.