Possibly redundant patchghost copies

[PhilipDeegan]

Currently in both solver_ppc and Ion_Updater::updateAndDepositDomain_ we copy the patch ghosts per patch during advance.

It's possible that one of these might be redudant.

Having tried removing the copy from solver_ppc, we see test failures in test_advance::_test_L0_particle_number_conservation

with particle count mismatch,

however if we remove the copy in updater, we only see test failures in test_updater.cpp @

TYPED_TEST(IonUpdaterTest, particlesUntouchedInMomentOnlyMode)

which might just not make sense anymore

the above can be seen on my fork @ https://github.com/PhilipDeegan/PHARE/pull/91