Simple Sorting Hoppers

Performace Improvments

Major

• Made it so a spatial chunk scan is no longer being performed after every item transfer attempt by storing filter status and last checked tick for every hopper.
  • Found this had ~95% reduction in CPU time
• Mutate the existing filter catche entry in-place instead of generating new one each time or re-inserting map keys
  • Removed Stuttering and random pauses
• Before there was the static 20 tick TTL so if there was massive hopper intansive setups there would be mirco spikes every single time, so I added a pseodo random offset (based on positional data) to spread them out.
• Stopped hoppers with no filters from running full spatial bounding box searches on every single one of the item transgers.
  • Removes all overhead for normal hoppers effectivly.
• Added bitmask check in hopper tick loop to prevent an O(N) linear traversal across all entries, causing frame drop.
  • Pruning is throttled using a bitmask check (currentTick & 0x7F) == 0, ensuring stale-entry cleanup runs at most once every 128 game ticks (~6.4 seconds).

Minor

• HashSet lookup for item search instead of .contains(stack.getItem())
  • O(N) -> O(1)
• CanAccept() was called twice per transfer so I switched it so the destination was not able to accept it immediatly canceled the operation.
• Prevented potential client-side rendering thread execution or level nullability issues.