Set the bits that aren't being used in the texel data to zero so that
the integer decoding sequence recreates the proper texel values for
trits and quints that don't have as many bits as there are expected
values.
I'm not completely sure what the best strategy is in this case. Ultimately, it's good
that the format itself carries the block dimensions. It makes a lot of the code somewhat
uglier though, but really the only thing that we're sullying is the succinct ability to
determine what large-scale format it's in (PVRTC vs ASTC instead of 2bpp PVRTC vs 4bpp).
Start with small tests to make sure that integer sequence encoding is working the way
that it needs to. Ideally there will be many more tests here, but for now we are trying
to get ASTC decompression working and testing is not a priority.
Fixed the notion that the CEM data was right after the first six bits
after the partition mode. It's actually right before the texel weight
data. Still not sure whether or not the texel weights are read in
the opposite order from the color data...
Also, added some preliminary integer sequence decoding. This will need
to be tested before we actually figure out if we did it correctly or not...
