Sustain is basically a measure of efficiency. You put energy into the string when you pluck it. Some of that energy goes in to causing the top to vibrate the air inside the soundbox, which we hear as sound. Some of the energy is lost to the instrument in that process, and the rate at which energy is lost is what determines sustain. So - if we want a lot of sustain, we want an "efficient" design - one that robs the smallest amount of energy from each vibration.
There are a lot of variables that go into efficiency, though. If we're talking purely about wood selection, then you can basically distill this down to stiffness (elastic modulus) versus mass. A stiffer, lower mass plate will lose less energy over time than a heavier, less stiff plate. Think of a sheet of rubber - really high mass and incredibly low stiffness. The worst! Luckily, wood is very good in this respect. And, no surprise, many typical soundboard woods are the best. Generally, softwoods have a better ratio than hardwoods - most fir, spruce, cedar, pine, redwood etc are very good. This is reflected in instruments more mature than the ukulele (i.e. guitars, bowed instruments) although some of the choices made in those instruments are out of tradition or at least what species were available when they were developed. Plain old boring douglas fir is probably the best among commonly available woods, but unfortunately it's associated with cheap 2x4s and it doesn't seem like anyone wants a uke made from a wood associated with cheap 2 x 4's.
In terms of the body, there are different camps of thought - at least, for uke sized instruments. Some builders try to think of the body as inert - you don't want it moving at all, you want the energy to all stay in the top. Other builders think of the body as a sort of sympathetic soundboard, so you want to build it efficiently, just like the top. Builders that use heavier, denser sides and backs (like rosewood) fall into the first camp. Rosewood sure is stiff but it's very dense. I don't think there's a right answer here, it's as much about the design choices as it is about the wood selection.
Which ultimately is probably the most important variable here - design choices. Not just choosing the right species, but also choosing the right piece of wood and making other building decisions. In my own building experience, I can tell you that it's possible to make two ukes from very different woods that sound close enough (including sustain, not just tone) that you can't tell them apart in a blind listening comparison. Choosing a specific wood in order to get to the goal of sustain can be important, but building choices are going to be just as important if not more so.
In terms of consistent volume up the neck - IME that's as much about string choice and setup as anything else. If I know someone is going to play up the neck often, I do build in a tiny amount of relief (versus building dead flat or with fall away) since that keeps the string break angle over the fret a little more consistent. Again though, there are lots of approaches to this and different builders will have different successful approaches.