If you embark on an incident light reading you have no need for a grey card.
That's true; but the essential difference between incident and reflected readings is that the former gives a direct measure of the amount of light falling on the subject, and the latter the amount of light reflected back from it. In the case of general reflected light readings, this means that if you measure the light reflected from a predominantly dark subject, there's less light than if the subject is predominantly light.
Incident light measurement only means that you measure the amount of light falling on the subject, and whether you put a white plastic dome in the light path or a piece of card is neither here nor there. Yes, incident light meters have plastic domes; but they could equally well have used a separate card (grey or otherwise) for all that it matters. So long as what you are measuring is subject independent, it's effectively
an incident light measurement. Which is all that I said. By measuring the light reflected from a subject of known reflectance, you are effectively measuring the light incident on the subject because the amount reflected is directly proportional to the incident light. A white dome will reduce the amount of light, so you're still not actually measuring the "neat" amount of light falling - you're still measuring a quantitiy that is directly proportional to the incident light but is not identical to it.
Incident light guards against overexposed highlights, which is the killer with digital and slide film; but for various reasons (to give them would swell this post far too much) it can result in underexposure (which is true even with digital and slide films) and isn't the best method to use with negative films where underexposure is the thing to avoid. The most accurate method is spot metering, which is a variant of a reflected light measurement.
To adequately go into all this would require a lot more space than is available in a forum post, and cover a lot of more or less technical background info.