Lifting a cabinet using actuators or scissor lift

In a scissors jack design, items one and three, you'll have the least mechanical advantage when the assembly is retracted. Using drawer slides or the equivalent is a good idea, but will not provide for a balanced lift if there are loads applied that are not uniform over the cabinet's bottom surface.

Your selected motor, with a thirty to one reduction should have plenty of torque. As you've not specified any desired speed of action, one must believe that slow is okay. Slow is to your advantage as well.

If you select the vertical screw design, which is similar to many 3D printer beds, you can get away with threaded rod. It's not used in 3D printers or other devices which require precision due to backlash and imprecise manufacturing, but for your application, you need none of that.

You'd have the advantage of being able to drive all four screws by using a cogged belt and pulley system, again similar to that used in 3D printers and in my laser cutter bed lift, keeping the screw lift synchronized during use. It may be necessary to have a more powerful motor or use a finer threaded lift rod.

Finer threads would mean more available lifting power, but much less speed. Keep the rod lubricated and clean and you would avoid the higher expense of the trapezoidal nut and rod design, common in 3D printers. (I couldn't resist adding that.)

You'd be able to fine tune the ratios by changing the diameter of the cogged pulleys on the motor and on the threaded rod to provide for faster lifting or greater torque/power as required.

For the belt around four pulleys, you'll get 90° wrap without tensioners/idlers/guides and if you add guides to the system, you can set up to about 270° wrap.

Your threaded rod may not be perfect, but you can run a die to clean it or even an ordinary nut multiple times to clear the burrs. Lubricate while doing this and it should run smoothly in short order. The nut will be easier to manage, in my opinion, as the die is designed to cut, unless you get a thread-chasing die. Thread chaser is on the left:

Even if your threaded rods have a bit of a bow or wobble, it won't affect the overall performance, again, the precision non-argument.

The motor you referenced in the comment is three times the torque and one-quarter the speed. If you change other ratios to make one equivalent to the other, the slower motor loses out in the equation.

You may be able to calculate roughly the lift rate once you know the thread pitch. That's relatively easy if you're using metric hardware.

Start with configuring a one-to-one ratio for the motor to the rods. 270 rpm is pretty fast. With a 1 mm pitch, you'd have the assembly climbing the rods at 270 mm / minute (~4 mm / second) for a (500/4) 125 second climb, slightly more than two minutes, roughly calculated. After doing all that intermediate math, I realized that 270 mm / minute and 500 mm climb is less than two minutes. Rounding doesn't always work well!

As in 3D printer design and laser cutter design, the tops of the rods should be restrained. Retainers could be as simple as a slightly larger hole in a plate. 3D printers don't use bearings, because it creates artifacts in the print, while laser cutters have bearings for the top mounts. When the cabinet is at maximum lift, the bearing at the bottom and at the lift point will restrain the rods, but at minimum lift, the ends are free to whip about.

I lack the engineering background to address the torque vs lifting power in this configuration. My method is TLAR* and trial and error.

*that looks about right