The 2N3055 SOA and 723 argument may well have merit.
30V is about the top end of what a jellybean opamp can take, and back in the day the entry point to the game was often audio projects of some kind which were usually either +-15V, +12V or +24V give or take, so it also fitted well with what people actually needed.
Reality is that almost everyone ends up with multiple power supplies in the lab, there is no one size fits all.
I have a couple of TTi/Farnell jobs that do the usual 30V at an amp or so, a couple of big switchers for more serious power, two HP programmable units, and a biscuit tin with two stacks of D cells and some BNC connectors for testing very low noise, high impedance things (And I am not shy about throwing an extra supply together if I have a particular need).
These days I do not consider a power supply to be a particularly good early project, you can find a second hand TTi or such for not much money, and a general purpose lab supply is actually quite tricky to do well, at least until you have some control theory and some experience of the practical stuff under your belt.
Regards, Dan.