These inclusions are anything but "simple", even in material set theory. Consider the real numbers; in common formalizations, a "real number" is an equivalence class of infinite sequences of rational numbers. You can't understand the 'inclusion' of rationals in reals as material (set theoretical) inclusion; you need an adapter/wrapper function. Type theory works the same way.
Quotients are not "natural" from a constructive POV, and that's ultimately where the difficulties you're referencing come from. If you're OK with being non-constructive, quotients can be formalized ad hoc. (One of the positive side effects of homotopy-theoretical foundations is that they might help clarify what exactly is going on when one reasons on quotients.)