It is not settled that there was a Celtic pantheon. Several hundred divine names are attested from Gaul and Britain, and the great majority appear once, at one place, often paired with a Roman god in an inscription set up by someone working within Roman conventions. From that material it is difficult to demonstrate a shared organised body of gods rather than a shared style of religion practised locally.
What can be said is that no single sovereign presides. A handful of names recur widely enough to look pan-regional, Lugus and the horse goddess Epona among them, but they do not form a court, and there is no succession narrative from the continental material at all.
The Irish material is much richer and much later, and it is the source of nearly every structure a reader will have seen attributed to the Celts. There the Tuatha De Danann arrive, defeat the Fomorians in battle, hold Ireland, and are in turn displaced by the sons of Mil and retire into the mounds. That is a genuine narrative of divine succession by conflict and settlement. Whether it can be extended to Gaul or to Britain is exactly what scholars disagree about, and the atlas records the claim as debated for that reason.
The pattern that does hold across the whole area is locality. Gods attach to a river, a spring, a hill, a tribe. The divine geography is the real structure, and the pantheon, if there was one, was assembled from it rather than imposed on it.