So my working theory and solution here is to vent to exhaust(basically atmosphere as far as emissions are concerned), and or use a good PCV cooler/condenser before allowing PCV gases into the intake. Then, occasionally use this CRC solvent for preventative gunk removal as needed (I'm hoping I can figure out a way to get a "borescope" to the intake valves without too much trouble). CRC could potentially also be used in place of media blasting, for direct cleaning.
I'd like to get the CRC injected as close to the valves as possible, so yes right before the throttle seems good. Ideally there would be direct ports in the manifold to inject this stuff right onto the valves - perhaps such service ports could be installed on the intake manifold.
I really wonder what the properties of this stuff are - I'm guessing it's highly volatile and will eventually vaporize and be combusted or expelled. However, if it were super volatile (think acetone), it wouldn't stay around in the intake long enough to soak the carbon for the hour that CRC suggests.
The outstanding concern with solvent seems to be letting it get down the cylinder walls, potentially causing some kind of related problem. For large carbon buildup, i agree there's a risk of material chunks breaking off and hitting the turbo. So, preventative treatment is definitely the way to go
PS: as far any near term testing I might do, I would hope to characterize the behavior of the CRC chemical(which appears to be patented), verify it's ability to dissolve carbon coking on a test sample, and perhaps see if I can "borescope" my valves. Assuming the chemical does in fact dissolve deposits well, then the rest of my testing will mostly be long term observation, which will take months and years to accumulate miles. My GTI currently is not my daily driver, either. If I set up a PCV cooler/condenser, I'll report on its effectiveness.