Tidy-up AssmScatr, NeuBndCond, and DirBndCond CUDA implementation

Issue/feature addressed

• The AssmbScatr kernels are updated for coalesced memory access
• CUDA grid size calculations are now encapsulated within kernel launchers. This will eventually allow grid size optimization for each individual kernel and for specific GPU device, if desired.
• Eliminate duplicate host memory allocation in operator constructors. Nektar Array's are now used for temporary host memory storage. Memory is deallocated after construction.

Proposed solution

Implementation

Tests

Suggested reviewers

Please suggest any people who would be appropriate to review your code.

Notes

Please add any other information that could be useful for reviewers.

Checklist

• Functions and classes, or changes to them, are documented.
• [ ] User guide/documentation is updated.
• [ ] Changelog is updated.
• [ ] Suitable tests added for new functionality.
• Contributed code is correctly formatted. (See the contributing guidelines).
• [ ] License added to any new files.
• No extraneous files have been added (e.g. compiler output or test data files).