There’s a lot more to designing a drivetrain than ‘more speed’ or 'more torque.
While it’s certainly true that, for the same gear ratio, a MiniCIM is slightly faster and a CIM has about 50% more torque, that’s not where the fun stops. It’s always possible to change the gear ratio to get the speed you want, and the torque you need. The choice of MiniCIM or CIM is usually driven more by thermal mass and packaging.
The calculator at VexPro (written by the great John V-Neun) offers a good glimpse into the basics of drive design. John has simplified it in a way that covers the basics fairly well. You will notice that there are cells for “Pushing” current draw per motor - multiply this by 6 to get the total for the drivetrain. What current are you at with CIMs? What about MiniCIMs? Do you think you can sustain that kind of current from an FRC battery, even for a short acceleration?
A number of years ago, I combined an analysis of the electrical system and some physical tests and concluded that 6 CIMs was unsustainable for power consumption reasons (brownouts, tripping main breakers, overheating battery cables, and batteries dying before the end of the match). Of course, this is dependent on the exact drive design, especially the scrub friction. 33 then went on to use 6 MiniCIMs as a compromise - Comparable power to 4 CIMs, but distributing the load across more motors for better thermal performance. Of course, YMMV.
I’d suggest you start by looking at the traction limited performance with both configurations. JVN’s most recent calculator will show you the pushing current draw - some of his older ones would also show you the pushing force (if your robot pushes a brick wall, how much force can it push with before the tires slip?). This force is constant - the mass of the robot times the coefficient of friction - but the current at which it occurs depends on the wheel diameter and gear ratio. If you gear too ‘fast’ (numerically low ratio), you will suffer very high current when launching and may experience brownouts or excessive battery drain. If you gear too ‘slow’, then you will reach top speed quickly and your robot will be slow as it traverses the field. It’s important to find the balance. Since you have a 2-speed, you could assume you start in low and then shift up, but that takes driver training. Make sure to factor in the possibility of a launch from a standstill in high gear.
A final thing to consider is torsional stiffness and scrub friction. Both are related and are important. Scrub friction is the friction the robot must fight to rotate, and is usually a function of the friction of the corner tires. If you build a drop-center drive, then the torsional stiffness is also important. A less stiff 6wd will tend to flex enough for the wheels on opposite corners to touch the ground, increasing your scrub friction. All of this is important when you select your gearing, because a robot geared for long field, straight line performance may be geared too high to turn if the scrub friction is too high. Again, you might have to shift down here if you’re geared too high.
tl;dr, gear ratios are very flexible and should be used to adjust performance instead of motor choice. Be careful with very high power drives that you don’t exceed the capabilities of FRC batteries.