Foucault pendulum movement.

Actually, a universal joint wouldn’t work.
What you were probably thinking of was a "snap swivel"

But, all of these have very significant friction - using one would cause the pendulum to be “slow,” I would think.

No, he used a universal joint. It worked very well.

The only reason for a motor is if it’s going to run continuously. If the museum staff resets the thing in motion every morning before opening then you’re probably right that inertia will carry it through the 12-ish hours the exhibit is viewable. That does however raise a requirement to be able to reliably and repeatably launch the thing every day using semi-skilled staff. How that’s done is left as an exercise for the reader.

Also see my link here The Nat | for a museum’s description of their large-scale pendulum exhibit which does have a motor.

I don’t see how - a universal joint does not allow rotation between the two shafts - only angular movement.

Much later add: This link Foucault Pendulum originally from post# 20 by rsa shows vids of a more-or-less museum scale pendulum which has lost about 1/3rd of its swing amplitude in an hour of operation. So it seems implausible an unpowered pendulum could last the full 8+ hours a museum would typically be open.