Kernels for scaling a vector by a diagonal matrix

[adhamsi]

Description

Implements HIP and CUDA kernels to scale vectors by diagonal matrices (element-wise product).

Closes #280

Proposed changes

Implements the kernels and a test that scales the incrementing vector of length n by itself.

Checklist

• All tests pass. Code tested on
  • CPU backend
  • CUDA backend
  • HIP backend
• Code compiles cleanly with flags -Wall -Wpedantic -Wconversion -Wextra.
• The new code follows Re::Solve style guidelines.
• There are unit tests for the new code.
• The new code is documented.
• The feature branch is rebased with respect to the target branch.

Further comments

[shakedregev]

Looks good and tests pass. Adding just one small clarification and merging.

[shakedregev]

Clarification: this also implements CPU kernels.