Follow up fixes for multiple muffler hatches.

[AbdielKavash]

This is a follow up to #3489 fixing several problems with that PR.

• Logic for pollution output no longer creates unnecessary copies of arrays, instead it picks a random index into the existing list of muffler hatches.
• Said logic is also simplified in case the multi has 0 or 1 muffler hatches (as a large majority of multis in a typical base will) to avoid hatch selection completely.
• Removed unnecessary array copy when generating info data.
• Info data now mentions and correctly displays average pollution output also for all other machines that override getInfoData.
• EBF and MEBF now calculate pollution gas output based on the average pollution output of all mufflers (this currently does not matter, as both require exactly 1 muffler, but it is calculated correctly just in case).