The ISS has always been designed as an integrated station of the Russian and US segment. There was never any need to duplicate functionality between both segments, as independent operation of each segment is non-goal.
It may be just down to the geometry of the station if the Russian sections are in the most efficient positions to mount the thrusters. Or maybe it’s because you need to have a single control system in control of each system, one for the thrusters and one for the flywheels, and it’s not viable to have those span the US and RU sections.
In practice you need both systems. You can’t just run up flywheels indefinitely, every now and then you need to use thrusters so you can spin them down. Meanwhile flywheels give you fine control that’s difficult to achieve with thrusters, especially on something as big and complex as the station.
Small quibble: the ISS gyroscopes are constant-speed; they apply torque to the station by pivoting their axes. The argument stands, though; if there's a constant need for torque then eventually the gyroscopes will end up all pointing in the same direction (a "singularity") and unable to apply further torque; thrusters will then be needed to balance the gyroscopes being returned to effective orientations.
