You wouldn't be "caving" on the "permits for everything" stance, would you?
First, let me say that I've only a GENERAL idea of what each decision you mention will do to/for you - your original intent was that your neighbors weren't annoyed (or even aware) that you were playing drums at any hour - so I went with the optimum isolation for each case, within your space constraints. Had this been a "from scratch" project, I would have recommended even wider gaps between walls, with everything else being similar to my other recommendations; this is construction I KNOW will work.
When you eliminate decoupling between inner and outer structure, you get flanking; this is such a difficult thing to figure that it's been known to make only a few dB of difference, up to sounding like the offending noise is being generated in the SAME ROOM - only safe way here is to decouple IMO.
As to doing a riser, that will work for impact and coupled sound directly from the kit to the floor; however, if the rest of the room is NOT also similarly floated, the airborne sound will transmit thru the un-floated section of floor and into structure. If you built your riser large enough so at least the bass amp also sits on it, that would help. Guitar amps could be set on MoPads or similar for a few dB improvement in isolation.
Your perimeter expansion joint, being of wood, will likely attenuate sound between slab and stemwalls/frame, but not STOP it; wood is nearly as good as concrete at transmitting sound thru shear waves, just at a different speed - for example
http://www.uk-piano.org/sound.html
Note the speed of sound in wood has Three different values, depending on direction - across the grain is around 4300 fps for Fir, discounting the possibility of travel ALONG the rings.
Concrete - average sound velocity in concrete is around 4,000 meters per second, or around 13,000 feet per second compared to sound in air at around 1130 feet per second;
So, sound traveling thru concrete/fir/concrete would experience two speed changes (and be diffracted twice) - this entire process would attenuate sound somewhat, but I'm not engineer enough to be more specific than that.
Bottom line - not floating your inner walls (I think) would be a mistake - I don't have the capability of specifying the way to do that, independent of being set on a floated floor - maybe Kinetics has standard ways/materials for this, seems like I saw something. Possibly build walls/ceiling floated, then test, then add a floated floor of some kind if/when it's not adequate.
If your walls are done first, you can always build an inner, floated floor later if it's not good enough; it's just not as easy as building from the inside out, things are harder to maneuver when there are already walls in the way... Steve