* add fibonacci bench https://github.com/status-im/nim-taskpools/issues/5
* unify allocs, don't use a mix of calloc malloc and wv_alloc
* Chase-Lev Deque: "unlimited" growth
* Remove affinity / CPU pinning support: does not work for ARM (Big.Little Arch), macOS, Alder Lake (P and E cores) and multiple instances of a program get the main thread pinned on the same core.
* Remove weave-specific things: WV_NUM_THREADS, the design-by-contract asserts
* avoid running destructors on freshly allocated tasks on Nim 1.6
Mamy Ratsimbazafy