For my own personal junk if something was designed to work with MySQL and would require code changes to work with Postgres I personally would stick with MySQL or Percona MySQL in my case rather than forking a project I do not maintain especially given I am not a proper developer. That's just for me and my silly hobby projects.
I've seen similar logic applied in corporations. Product / Project management would have to agree to allocate development cycles or story points to maintaining a fork of a third party platform. Even if it's just a small amount of code someone has to babysit it and factor it into QA tests. Whether or not this makes sense would vary wildly by each corporations needs, core competency and what relationship they have with the third party platform maintainers.
* It works out of the box, and continues to work with no (or only minimal) tuning (no vacuum, etc)
* Scaling up through replication is straightforward
* Every techstack supports MySQL/MariaDB
In short: Good enough, No drama.
In my experience developers prefer PostgreSQL, while SysOps / DevOps / SRE's prefer MariaDB.