This is pretty uninformed, both about science and about f.lux.
If you want to talk about sleep and alertness, you don't talk about 443nm light, you talk about the "blue-green" peak response of melanopsin (~488nm, and a wide range around it). They are really far off here.
f.lux has offered "soft" dimming (which helps with PWM) for 4 years now.
They claim to reduce color temperature to 4500K which removes <25% of the effects of bright light. Of course you can use f.lux to do this too - you can choose any setting you like.
It seems like they test displays for PWM, so that is nice because you may not want to buy a display that has it.
But once you do, it's pretty hard to avoid: many models will make PWM when not dimmed at all, or the power settings will dim the display slightly, all the time, or Intel's drivers will dim the screen automatically when you use a profile like the ones they suggest. And as I said, f.lux's dimming will be equivalent in any case.