First 5 blocks are reserved for file accessing info, but we should include them when checking length
Closes#3468
Capture the bool indicating Japanese for the save file, so box names are now read correctly (if changed from game's default).
Getting close to weeding out all the easy inefficiencies to make me finally tackle the "stop gathering all moves instead of checking individual moves" allocation issue.
net6 was 4x faster than 461 with the old implementation; now it is faster for both but not as disparate.
New implementation performance is always at least 30% better than the old implementation; even better for smaller chunks of data (40% less time).
Pretty much mimics bare metal value writes
Sweet Heart still unobtainable (why did they give it a sprite if it's not functional ingame?)
Azure Flute is now obtainable
Sacred Ash was never obtainable
ddc6f414df caused it to be `<=` instead of `<`, but maybe the underlying issue was fixed a different way since all tests still pass with this reversion.
* Reuses move parse result objects for each encounter parsed in a LegalityCheck attempt, instead of creating a new object.
* Ensures the objects are never-null, and makes cleanup easier.
Slightly adjusts some other parts of the moveset validation to reduce allocations.
Have `Shiny.Random` be `0`, so we can skip init on this field for EncounterStatic. Plus makes it a little less brittle for future expansion if shiny qualities change.
ALM runs each legalization attempt in a separate thread, and if things are fast enough, we get multiple threads with the same Environment.TickCount seeding their Rand generator, resulting in the same PID/EC if similar templates are used.
