I can understand this from an executive decision making space ("well, we don't expect the Azure SQL Edge users will make full-fledged TSQL users, we're just selling more Macs"), but not a technical one, given aarch64's preeminence in the IoT space, and the fact that it would make sharing data in the enterprise trivial for Azure/MS shops.
Regarding the suggestion of running a full Windows VM or depending on an AMD64 container: Rosetta 2 support may only be available for a limited amount of time on macOS as well.
It would seem Azure SQL Edge was a nice experiment, but giving up would be ceding the space to PostgreSQL, et. al. I really liked the functionality of Azure SQL Edge, but I've already downloaded and configured PostgreSQL to my liking. However, I hope in the future MS tries other bold, interesting and useful experiments.