for many years in building motorcycle race motors, I lapped the flywheel to the crank taper.
Greg,
what mm wrote is pretty much your only choice. The shaft is probably ok, it is the inner hub surface that is deformed. That is more difficult to address but not impossible. Lap with fine (320 grit) tapping compound.
I worked at a ge apparatus service center for about a year. One major part of our business was to rebuild diesel locomotive traction motors. (600hp dc) the motors had a drive pinion gear say 10 inches in diameter and 8" long that was fit on a tapered section of the armature shaft. The fit was checked with dykem and hand lapped on the high areas using long strips of croakus cloth (shoe shine style). And re-witnessed. This went on for hours until a contact pattern was achieved that met pre existing standards.
The taper interface had to be absolutely free of any and all oils and residues. Mek and the like was used over and over maybe 10 times with special towels until we/they were happy it was clean.
The pinion was then induction heated and jacked into place flush with the end of the shaft. No other mechanical means of retaining the pinion were employed only the taper.
That is the long and short of it..
The torque a "dry fit" taper will endure is a function of only two things.
1) the taper interface clamping load or (determined by amount of stretch the ring (hub, pinion) is subject to.
2) the coefficient of friction of the interface (materials and surface finish)
surface area really is not a factor in the calculation. However (always a however)

the contact pattern must be wide enough to enable a stable fit and not rock.