It's really wrong, because we _should_ be storing PC + 2 in the stack.
And software's definitely gonna bork when we try emulating them, because
they will invariably inspect the stack and assume that's what we have in
there.
But the proper fix is to not actually do next_byte(), and to never
advance PC outside of the execute function. But that's a bigger change
than I want to do at this specific moment.
So, in the meantime, you have me prattling on in a commit message. Lucky
you!
This test broke once we began to (properly) push P onto the stack.
However, I'm not _clear_ on whether we should push P before or after we
set the INTERRUPT bit... We'll go with things as-is for now.
Criterion's init and fini config options allow us to register setup and
teardown functions. We now use this to register the common cpu variable
for the mos6502 tests, albeit through the use of a global variable.
Doing so necessitated that each of the different test files have their
own suites, due to some implementation details in Criterion, but this is
No Big Deal.
