Even a really low-power fission reactor puts out a few megawatts. A few MW of photovoltaics, even in Earth's orbit (and it only gets worse as you head out to Mars) is huge and fragile. Ion engines (or any other form of electric drive) are extremely energy-hungry; the energy demand goes up as the square of the exhaust velocity (E = 1/2 m v^2) and the whole point of electric drives is that they derive their extreme efficiency in terms of reaction mass by using absurdly high exhaust velocity.
There may be a point where it's practical to run a spacecraft propulsion drive (rather than a satellite's station-keeping thrusters) using electric thrust, and yet the energy demand is low enough that it makes sense to use photovoltaic, but I'm not holding my breath. Developing a drive practical for interplanetary flight would probably require literally orders of magnitude more thrust than currently-flying ion drives produce, so the fact that those can get by on photovoltaics really shouldn't be taken to mean anything useful.