The task db:reset runs db:schema:load instead of db:migrate. Written that way, any people using FKs or initializing data in migrations have to switch to the brand new db:migrate:reset. That smells like premature optimization, which is the rationale? Wouldn't be easier to let db:reset run migrations?
FKs are not supported by the framework itself, and data initialization in migrations may be said to be bad practice. In that sense db:reset is coherent and I don't open a ticket because of it. But by almost the same price you could avoid interfering with those practices. At least FKs is a common one.