- Chunk loading now is done in a more simple manner. Instead of splitting it into regions and having one file handle for all chunks in the same region, now all chunks make their own file handle for the region, the previous overhead was not worth it, since this seemed to increase performance.
- Speed up of ~15%
- Now the `.read_chunks` function does not return anything, instead it sends event to an mpsc channel.
- Dramatically decreased memory footprint.
Cause:
The size that is given in the file for the length of the chunk includes the compression byte, thus when reading a chunk that is at the end of the file, it would read 1 byte more that should be possible, causing a panic.
- Load chunk (0,0) only in debug mode. This prevents crashes when the chunk for (0,0) does not exist, since this is only used for checking the size of the chunk packet.
- Remove log: `Getting chunks x, from region file y`
