Use Array3 in HPMG solver

[AlexanderSinn]

Based on #1326

This PR replaces amrex::Array4 with Array3 as it has slightly more efficient indexing. Additionally, some tweaks are made to bottomsolve_gpu to get a speedup by solving the two independent components of system type 1 in two separate GPU blocks.

nx      speedup
255     16%
256     19%
1023    9%
1024    9%
4095    2%
4096    2%

• Small enough (< few 100s of lines), otherwise it should probably be split into smaller PRs
• Tested (describe the tests in the PR description)
• Runs on GPU (basic: the code compiles and run well with the new module)
• Contains an automated test (checksum and/or comparison with theory)
• Documented: all elements (classes and their members, functions, namespaces, etc.) are documented
• Constified (All that can be const is const)
• Code is clean (no unwanted comments, )
• Style and code conventions are respected at the bottom of https://github.com/Hi-PACE/hipace
• Proper label and GitHub project, if applicable

[MaxThevenet]

Thanks for this PR!