The GPUs you quoted, are not part of RX 4XX lineup. They are R7 4XX line. Polaris has its own branding starting with RX 4XX.
I should have typed Rx 4XX (where 'x' is the wildcard to mean over the whole range. ) I don't think AMD intends for "RX" to cover 'top to bottom'. If they do then there is tons of empty slots at the moment.
About the time span, it is a good question. The fact that they have Polaris line going to be rebadged shows that this is what AMD offers for low end and mainstream markets, however.
Nothing points to non iGPU Vega being pointed at dGPU mobile. All of AMD's slide decks on Vega don't particularly say much of anything about major mobile issues. In my book, mobile is part of "top to bottom".
Vega 10 - high-end
Vega 11 - Mid-Range(GTX 1070/1080 level of performance).
This year we are supposed to see some of Ryzen CPUs with GPUs. They are going to use Vega architecture, so it would account for "sort of top to bottom launch"

.
Vega 11 is mdrange or iGPU Vega. So far Vega 11 is missing in action in leakage into most drivers (last I did a scan). Again this goes to "how much time". Intermediate term it looks like Polaris is holding up the "affordable" mid range.
You have to use 12 memory cells to get 512 GB/s bandwidth, with GDDR5X. To get the same effect with HBM2 you have to use just 2 memory stacks

. Do you think manufacturing costs of HBM2 are 6 times higher in this particular case?
This is just misdirection. the 1070 is in the 200GB bandwidth range. Likewise for Polaris. Yes, if want to go to the very high top end range HBM2 is a better back for the buck. The real issue i that GDDR5 (and somewhat soon GDDR5X) is playing the same role that DDR3, GDDR3 , have played in the lower half of the line up range. Mature tech that uses much of the mature fab and design infrastructure that is already paid for.
HBMv2 is on the other end of the maturity spectrum. No long term high volume track record. Relatively new manufacturing technology that not been paid for already. etc. etc. Yes, HBMv2 is less power but it also comes at substantially higher cost at the moment too. solid "mid-range"
and affordable just not its wheelhouse right now.
The base GPU for Mac Pro could very well be Polaris 10 XTX/XT2/whatever_its_called.
All the more why it is a bit nonsensical for them to block the whole product line because the a much smaller percentage of very top end GPUs might be blocked. Release a Mac Pro and then over time release some new GPUs to go on time. Actually don't hide in rip-van-winkle mode for years at a time.
When they back themselves into a corner is where there has to be be new Thunderbolt
and new GPU
and new CPU
and new SSD it jacks up the product roll out.