Python implementation of the ALP (Adaptive Lossless Floating-Point) compression codec for double-precision arrays. Bit-exact compatible with alp-java
