It basically just revolves around well being, and consequentialism (how each action or inaction affects the individual and everyone else) although it's bit more complex than that.
And it's not a perfect system, there are limitations. But I don't think a rigid set of rules suits ethics and morals.
For instance, is it always wrong to steal an apple from a store? I can dream up scenarios where it would be the best thing in the world to steal that apple.
What about stabbing people with sharp objects? Is that always wrong? Well not when we're vaccinating our kids, or when granddad needs open heart surgery.
I like the philosophical idea that we should try to design a society in which we ourselves don't know who we'll end up being in that society. Rich, poor. Male, female. Black, white. Disabled, abled. Gay, straight, etc. That's a good way to make it fair for as many people as possible.
As for your example, killing one person to save many. They've done this philosophical exercise in universities. And it turns out that most people would kill the one person to save the many, if it's done remotely, by pressing a button or something. But far fewer will if it involves them physically killing that one person (pushing them off of a bridge, or strangling them or something).
You're right. That one person though could solve global warming. But they could also be the next Hitler. So I give them a sort of neutral position on the Hitler scale. There's no telling what lies in their future. Probably nothing extreme on either end of the scale. So then you're left with the probability of all involved being somewhere in the middle of the Hitler scale. And it becomes mathematical. 1 person VS say, 10 persons.
I'm going to take that test and I'll post what I find.... The test that you posted, not the test I alluded to above. :P