Before the advent of TTL-flash the standard way of doing field-macro was to have one bracket with a flash and another bracket with a white reflector, or two brackets with flashes, or some aluminium foil around the lens and making a reflector on the side opposite the flash.
The flash would always fire at full power because light is never too much and DoF is never enough.
Forget diffraction: DoF for macro is better than absolute sharpness. F/22 is normal.
The right aperture would be determined by tests, and the closer the better if you have a setup that allows to maintain automatic diaphragm closure;
The focus plane would be fixed, as in the field it is a bit risky to recalculate flash position, reflector position, aperture etc.
So you would have this kind of "gun" composed by camera, lens, brackets, flash, reflector, set for a certain fixed magnification ratio, such as 2:1 and then you would go in the field hunting subjects always taking pictures at 2:1 and focusing by moving the entire gun.
In the field, you would concentrate solely on subject and focus.
You would preferably use a tele lens if the interest is for insects, as those, unlike plants, have this nasty habit of flying (or jumping) away just between the moment you focus and the moment you are taking a picture.
EDIT If this message seems to have no purpose, the purpose is to suggest to forget flash auto-exposure with a separate sensor. Those exist (at least Panasonic IIRC made a flash which could accept a separate thyristor sensor) but waste flash power and DoF. The double bracket solution prevents taking pictures at different magnification ratios in any case, so for a certain field setup the focus is "fixed" and better it is to use flash at full power using a pre-determined small aperture.