Currently Tracy assumes that users will be posting screenshots of frames via the tracy::Profiler::SendFrameImage API. Because of this Tracy starts an additional thread and includes code for performing on-the-fly DXT1 texture compression. Since we never need this it'd be nice to be able to toggle all of that off to reduce the code size impact and extra runtime threads.

HLS often uses arbitrary precision arithmetic. In this model, most operations (e.g. add, multiply) have wider results than their input, in order to represent all the possible result values without overflow. Explicit Truncation operations are necessary in order to limit the growth of values. The rules are roughly:

bitwidth(a + b) = max(bitwidth(a), bitwidth(b)) + 1
bitwidth(a - b) = max(bitw

Since #384 we can make use of the IR verification features of MLIR.

Our BConv op has a few parameter combinations where it can throw during Init or Prepare. In the converter we try to make sure that none of those cases will ev