Release v0.2.0

The release introduces a single library for both CUDA and CPU/Intel GPU, and all the major analytical patterns.

What's Changed

GroupBy with baseline hash table modification by @bagrorg in #16
Add plain join dwarf by @kurapov-peter in #17
DPL kernels build fix by @kurapov-peter in #21
Update dpcpp compiler by @kurapov-peter in #24
Fix deprecated namespace by @kurapov-peter in #25
Remove slab test from ctest list when experimental is disabled by @kurapov-peter in #23
Hash build with non bitmask table by @bagrorg in #26
Slab hash final implementation by @bagrorg in #15
- tbbsort and groupby with local hash tables by @bagrorg in #28
Add multi-stage results and fix groupby file format by @kurapov-peter in #30
OmniSci Join dwarf by @bagrorg in #31
global_ptr issue in comparison fix by @bagrorg in #34
Dwarfs expected checks and HashBuild fixes by @bagrorg in #36
Temporarily move omnisci join to experimental and use default dpcpp c… by @kurapov-peter in #38
Fix Dockerfile with additional open-cl driver by @lmontigny in #40
Fix NDEBUG in tbbsort by @lmontigny in #42
Fix data type for CSV file by @lmontigny in #43
Update README.md by @lmontigny in #47
Add initial installation scripts by @kurapov-peter in #46
Update nvidia GPU support by @lmontigny in #49
Experimental build issue fix by @bagrorg in #51
Dwarf bench library for integration by @bagrorg in #50
Fix GPU execution for Dwarfs and minor API fixes by @bagrorg in #56
Update docker file to use all the components direclty by @kurapov-peter in #57
Add ocl installation for cpu in dockerfile by @bagrorg in #58

Full Changelog: v0.1.0...v0.2.0