Refactor code CUDA and Native backend to match #autumnai/collenchyma/62 that
provides enchanced memory management and syncronization. Since memory
management is now automatic, `*_plain` variants of functions are removed.

BREAKING CHANGE: *_plain versions of API functions are removed, arguments of
their counterpart functions may have changed in mutablity.

REFERENCE: autumnai/collenchyma#37, autumnai/collenchyma#62

Fix tests and benches to work with new memory management API. Refactor
functions for writing data to tensors and checking their contents. Merge
set up stage and compute/verify stage: it's much nicer to see input and
output values in one place. Refactor tests to be generic on backend and use
macros to eliminate repetitive test definitions.