Good discussion!
Mass flow maps are the exception, not the rule. They are typically presented as a convenience to the customer who wants to compare them. They ALL begin life as volumetric axis maps, then a standardized conversion factor is applied to give them an apples-apples usefulness . Hence the MAP is pretty much invalid except for ONE unlikely set of conditions. This is also why they are not a very good absolute reading of actual mass flow rates, until it is corrected for non-standard conditions.
Centrifugal air compressors are not mass devices. They have no idea what the density of the air is.
...and they don't leak, they just have a limit of "push" capability. They are constant head, vs constant volume. But they will spin harder/faster to overcome restriction to satisfy boost set point...they "push" harder. Hence all VGT are subject to overspeed, and there is no safeguard or upper limit except failure (unlike a wastegated turbo)
Restriction is the mortal enemy of the VGT turbo compressor, especially suction side restriction. Few things matter more. I am confusing some, "variable displacement" and "constant head" are different ways of saying the same thing. Sorry. For the compressor, "variable" results from several factors, the most aggregious one is restriction from design plumbing. But then that applies to all centrifugal type "constant head" pumps.