The calibration prints are separate for PLA with 0.95 and PETG/ABS/ASA with 0.98 extrusion factors. When both are used, deviation is a bit more present but from my point of view negligible.
I don’t know if it would work, but with what you describe, I would only use flushing (no prime tower) but place a fitting geometry close to & around my object instead. – A good old solid wipe shield that also minimizes stringy distances.