Modifying lcmgen doesn't rebuild the tests

[judfs]

Say I checkout master and run make; make test. -> "100% tests passed, 0 tests failed out of 14".

Now I add some garbage to emit_python.c and run make; make test. I still get "100% tests passed, "

But if I run

rm -r test
make -j > /dev/null
make test

Now I get the expected "13/14 Test #13: Python::client_server ...............***Failed 1.69 sec".

This seems super bothersome to fix in the cmake... At least from what I can tell. Might be best to just document this in docs/content/contributing.md?

[nosracd]

This seems super bothersome to fix in the cmake... At least from what I can tell. Might be best to just document this in docs/content/contributing.md?

I briefly looked into a programmatic solution but nothing jumped out at me, so I'm thinking documenting it is the correct solution. In that case, the issue is resolved by #507.