I think you're right, but I think this doesn't address what I suppose is a reasonable fear: engineer a significant enough advantage to out-compete a rival population, and then only later rediscover the need for diversity. That doesn't work out well in the long term for either population, but that doesn't mean that, given the choice, it wouldn't happen.
Given the Chinese example, I'm immediately reminded of the folly of The Great Leap Forward and Chinese steel production quotas, and their results.
Reasonablish, although we do (and I feel a bit monstrous saying some of this) already have existing models in the form of NA reservations and the Amish.
Both of these are (massively?) out-competing groups (one of them through "no fault of their own") who are maintaining existence.
Even a bit more of a deliberate effort on the part of the competitive majority to support any (and again, feel a bit monstrous) "baseline reserve population" would probably succeed enough.
Enough people living today will have their genome sequence that if that's true you can just edit in the 'diversity' later in the future if that's needed.
Given the Chinese example, I'm immediately reminded of the folly of The Great Leap Forward and Chinese steel production quotas, and their results.