If you look on the Sharkscope leaderboards for the 2013 year,  the majority of the list of top winning players is comprised of hyper turbo regs (some of whom  play other formats as well but for most of them HU hyper turbos constitute the majority of their 2013 Sharkscope winnings -- you can verify this by filtering by game type for Heads Up Super Turbos). 
Yes, in my opinion, this format is well worth learning. Looking over the Sharkscope leaderboards, it's pretty clear that HUSNGs are the way to go -- it's bloody hard to make a living wage in many of the other SNG formats.
Suppose you play $7 hyper turbos and have a 4% ROI. Then you make a profit of $0.28 per game on average. Assuming you can put in (conservatively) 15 games per hour one tabling, that means you're making an average of 15 x $0.28 = $4.20 per hour. Two tabling the $7 hypers with a 4% ROI would yield $8.40 per hour. If we played 40 hours per week, that'd be an average profit of $336 per week =~ $1300 per month 2-tabling the $7 hypers.
A 3% ROI at the $15 hypers would yield a profit of $0.45/game. Assuming the same 15 games per hour, one tabling would yield $6.75/hour and 2-tabling would yield an average hourly of $13.50 -- that's $2000+ per month you could make by 2-tabling the 15s with a 3% ROI for 40 hours per week [playing 20 hours per week at the 15s 2-tabling with a 3% ROI would be a nice way to ''top up'' what you make at your full-time job by $12,000/year]. 
As for your other questions... 
It's probably best an American (eg. not me) comment on the legality issue.
Software isn't necessary to achieve the kind of results I've mentioned above. I'm sure a hyper reg could better comment on what effect a potential ban on applications like sharky would have at higher stakes.
I doubt the games will get drastically harder at the micros. It's probably fair to assume that the higher stakes will continue to get more difficult (eg. how many winners at the $100+ level are going to quit playing poker when they make more playing poker than they would doing other things? --> Unsurprisingly, we'll presumably see more and more regs)
I see hyper turbo HUSNGs continuing to be very popular and beatable at the lower stakes. I envision it becoming increasingly difficult to be a winner at mid stakes+ for the reasons I've previously mentioned. 
All just imo. I'm just a $7 donk tho :(

Nobody will be able to tell for sure. There have been times, where FL FR Cashgame has been the most played games ever. NL CG took over. Then FR died and SH did dominate. There have been times not lot of players did care for HU SNG, then fish did discover them and so did more and more regulars.
Maybe there will come up a new poker variation that will kill off the traffic at HU SNG. I believe that tools like Sharkystrator did make HU SNG bit more profitable again. On the other hand, it is now much easier to hunt down the fish and the sea may be empty sooner.
The only thing that can be said for sure, is that when you learn HU SNG you have to "master" some skills that you can use in other forms of poker as well.

Whats the max multi tables that people can comfortably play for HUHypers? I tried doing 4 and it was a little brutal.... I use to play 30 9man turbos fine, but HU hypers is crazy... 

Learning one format of poker will help you a lot in others areas of poker as well.
Many disciplines are the same no matter what game you play:
- tilt control
- observing opponents tendecies and adjusting your game
- thinking in terms of ranges
- comparing expected value of different lines
I like to think HyperTurbo in poker is like chess endgames. Every serious student of chess  spends some time studing endgames. Every serious poker player should probably spend some time playing Hypers.