Is this for serious use or a range toy? Not disparaging range toys, I have them. It affects recommendations for optics.
The remainder assumes you realize the independence of your two sighting systems, if not I can clarify more. Also be aware that different companies represent Glock rear sight heights differently: either from the top of the slide (like Dawson, I think) or from the bottom of the dovetail (Ameriglo). The dovetail is 0.079" deep, plus or minus (not sure what actual spec is).
On the sights question, you will probably need to do some research and testing on your own. Even purchasing generic suppressor height sets from Ameriglo (GL-429 set) I found POI way off POA where I wanted to zero, so I had to mess with individual sets. The first thing will be to decide on the optic and milling depth, and any adapters needed. From this you can determine "deck height," or the height of the optic's body above the slide's top surface. This changes with different optics, the RMR body is thinner than the DPP, for instance.
You will need to decide how much iron you want to see in the window, from "just the tip" so you can see just enough of the irons to get a reasonable POA, towards having the top of the irons in the middle of the window when aligned. If you define X% cowitness as "top of irons are at X% of the window height, and assuming you align the dot at 50% of your window height when zeroed and presented, I think my cowitness is about 10%. I originally had about a 30%, and I wouldn't go over that as you start to lose one of the dot's advantages, which is not covering up what's below your POA. Going too low, you start to lose the light between front and rear sights...so a 1% cowitness would have you struggling to align a tiny slice of sights... I think of the traditional "absolute cowitness" as a 50% cowitness in my world, where the "lower third" is a 33% in my terminology. Again, they are independent, it's really about where you want the irons in your window when the irons are properly aligned.
So once you figure that out you need to start looking at sets of sights that will give you the correct POI at your desired zero distance. I'd go with black serrated front and rear, as dots are confusing with dots, and you're also limiting yourself to seeing more irons in the window, else your iron dots are useless.
For reference, I had
ATEi mill my Gen4 19 for an RMR with sealing plate. They are widely regarded as one of the gold standards in milling work. Not saying there aren't other companies that are as good or better, this is just who I chose to go with.
I'd have to go back and check my notes to see all the measurements and garbage I needed to go through, but the long story short is the original recommended suppressor height sight set (
GL-429) had 0.315"F and 0.394" rear, and shot high at 10 and 25 (my two reasonable zero distances), and from memory was like a 30% of the way up the window. Based on my calculations, I figured I needed to change 0.035", either taller front or shorter rear to zero at 25 yards. This left me with the closest standard sights being a 0.360" rear (
GL-408) with my 0.315"front, or a 0.350" front (
GST-350) with my 0.394" rear. The first choice (keep front, lower rear) put the irons lower in my window (desired) while still giving me enough notch visible to shoot the irons, whereas the second choice (keep rear, raise front) put the irons higher in the window, more than I'd want, so I ordered the 0.360" rear and everyone lived happily ever after.
I can take pictures of the new setup and dig up pics of the old if you want, but my slide looks like anything off the ATEi gallery page: they reprofile the serrations (wider troughs, narrower peaks, and they leave them SHARP), add front serrations, add the fish scale top, and mill for the optic. Their pro cut is milled for YOUR specific optic, so the optic is press fit down into the cavity, basically puts all the loads on the optic housing rather than the little lugs.
edit: Doug at ATEi is a smart dude, I would reach out and talk to him, and listen to his advice. If he won't mill slide X for optic Y, it's probably a dumb idea to do so.