Owner occupancy covenants are the part of ADU rules that rarely gets priced in, so how should they be modeled?

Read three different ADU ordinances in the same metro and the thing that keeps changing the math is the owner occupancy condition, well ahead of setbacks or size caps. Two of the three might require the owner to live in either the main house or the ADU as a primary residence, recorded as a covenant on title. The third has no such requirement at all. What that does to the numbers: build a $180k detached unit renting at roughly $1,600 a month and the income covers a good chunk of the mortgage while the owner lives there. Fine. But the exit changes shape. A buyer who wants both units as rentals cannot legally do that under a recorded owner occupancy covenant, so the buyer pool is owner occupants plus people willing to move in. That is a narrower pool than the appraisal comps suggest, and appraisers do not appear to be discounting for it. The other wrinkle is that these covenants sometimes survive a zoning change. The city liberalizes the rule later, but the covenant recorded against the parcel is still sitting there until someone files to release it, and the process for releasing it may not exist yet. So the question for anyone who has read more of these: do you treat an owner occupancy covenant as a real haircut on the value the ADU adds, or as a paperwork item that clears at closing? The careful position is to model it as a haircut and be pleasantly surprised. Also worth asking whether anyone has seen a lender flag one during underwriting, since the expected pattern is that title picks it up and the loan officer shrugs.