If you truly have a photoresistor, then resistance as a function of brightness makes more sense. It's the resistance that is roughly constant for a fixed brightness as voltage is varied. Put another way, the current is roughly proportional to voltage at any one brightness.
However, if you actually have a photodiode or phototransistor, then the current is roughly constant at a fixed brightness as the voltage is varied. In that case, it does make more sense to plot current as a function of brightness.
It gets even more complicated. The above was assuming the photodiode was used in reverse mode. In other words, you apply reverse voltage to the diode, and the amount the diode "leaks" is proportional to brightness.
Photodiodes can also be used in forward mode, as in solar cells. In that case, they create their own voltage, and current at the same polarity as the voltage so as to produce net power. Neither the open-circuit voltage nor short-circuit current are all that linear with brightness (although the current is closer). The maximum power point shifts in both voltage and current as brightness changes.
You don't have to worry about forward mode with phototransistors. Those are effectively a photodiode in reverse mode amplified by a transistor, but all integrated into one device. There isn't a separate diode junction connected to a separate bipolar transistor.