I find that a bit off target.

It implies the reason for the change is because RBF is useful (which it is) and 0conf is dangerous (which it can be, although risks can be managed but may be significantly harder to mitigate the risks.)

The real reason was some Lightning Pinning attacks. A fix was suggested, however the mempoolfullrbf seemed to be necessary to solve the problem, at least that's what Lightning devs were able to sell Bitcoin Core devs on.

Here's a link to the Lightning dev mailing list showing the progression of the discussion within their community: https://lists.linuxfoundation.org/pipermail/lightning-dev/2020-April/002639.html

Also, highly recommend Bitcoin Explained, episode 68 on the subject: