Fix ME output hatch void protection again

[serenibyss]

ME output hatch void protection had 4 different issues at different points:

1. Mixing ME output hatch + normal output hatch on a DT (or dangote, MDT, sparge tower) can't run with fluid void protection on
2. Multiple ME output hatches, or ME output hatch + unfiltered normal output hatch breaks fluid void protection fully
3. ME output hatch can only hold 1 fluid according to void protection, requiring multiple for multi fluid output
4. ME output hatch can void some but not all fluids of a recipe that outputs multiple fluids at once

ME output hatches gaining a limited capacity introduced bugs 1, 2, and 3.
#3585 fixed bug 1, but introduced bug 3. Bug 2 and 4 remained unfixed with that PR.

This PR fixes bugs 1, 2, and 3:

1. Fixed by the checks in VoidProtectionHelper in the second commit
   • By testing for an ME output hatch, we can check if some amount of fluid can be stored in this hatch, and proceed as usual if so. Previously, this code would always view an ME output hatch as not being able to hold any fluid
2. Fixed by the check in MTEMultiBlockBase in the second commit
   • By checking if the tested hatch is an ME output hatch, we can see if this hatch has room to hold a fluid. If it does, we can safely insert any amount of fluid into it. If it does not, then we skip this hatch to test others. Previously, this code would blindly fill an ME output hatch, as a call to fill() on these hatches always claims to accept all fluid, regardless of if it actually can due to the cross-over between forge fluid api and AE2 api
3. Fixed by reverting Fix ME Output Hatch void protection checks #3585
   • This PR conformed ME output hatches to normal void protection rules. However, this code is not written in a way that allows a single hatch to store multiple fluids
4. Not fixed by this PR
   • Fixing this I believe will require a more thorough rewrite. Basically, void protection assumes that an ME output hatch can hold any amount of fluid, since ME output hatches can be "overfilled" to try and avoid voiding fluids. However, when a recipe outputs multiple fluids, void protection makes this assumption, but fill logic tries to fill the hatch multiple times. This can lead to some of the later fill calls not being able to fit, as the hatch is not able to be overfilled by multiple fill calls.
   • Solution for later: Implement a transactional system to fill the ME output hatch with multiple fluids at a time, OR rewrite void protection to not need to make this assumption with ME output hatches

Closes GTNewHorizons/GT-New-Horizons-Modpack#17711