I added fiber, made it noticeably worse, and only later worked out that I had added the bulk without adding the water.
Marking this solved because it was, and putting the answer at the top so nobody has to read nine replies to find it.
What it turned out to be: The mechanism is slowed transit plus reduced intake, and that combination determines which interventions work. Less food means less bulk, slower motility means more water reabsorbed, and the result is a smaller, drier, slower stool. Insoluble fiber adds bulk to a system that is already moving slowly, which is why it frequently makes things worse. Soluble fiber holds water and is the better first choice, and an osmotic agent works with the mechanism rather than against it.
So the question, as narrowly as I can put it: whether fiber actually helps GLP-1 constipation or whether the evidence is borrowed from a different population. Tell me what I have not thought of.