[EPIC]: Add Hopper features to `cuda::ptx`

[ahendriksen]

Is this a duplicate?

• [X] I confirmed there appear to be no duplicate issues for this request and that I agree to the Code of Conduct

Area

libcu++

Is your feature request related to a problem? Please describe.

Many Hopper features are not yet exposed through cuda::ptx.

Describe the solution you'd like

Add PTX instructions to cuda::ptx.

Describe alternatives you've considered

No response

Additional context

No response

### Tasks
- [ ] https://github.com/NVIDIA/cccl/issues/673
- [ ] https://github.com/NVIDIA/cccl/issues/1079
- [ ] https://github.com/NVIDIA/cccl/issues/1077
- [ ] https://github.com/NVIDIA/cccl/issues/659
- [ ] https://github.com/NVIDIA/cccl/issues/1339
- [ ] https://github.com/NVIDIA/cccl/issues/1343
- [ ] https://github.com/NVIDIA/cccl/issues/1344
- [ ] https://github.com/NVIDIA/cccl/issues/1350
- [ ] https://github.com/NVIDIA/cccl/issues/1398
- [ ] https://github.com/NVIDIA/cccl/issues/1399
- [ ] https://github.com/NVIDIA/cccl/issues/1400
- [ ] https://github.com/NVIDIA/cccl/issues/1401
- [ ] https://github.com/NVIDIA/cccl/issues/1402
- [ ] https://github.com/NVIDIA/cccl/issues/1365
- [ ] https://github.com/NVIDIA/cccl/issues/1439
- [ ] https://github.com/NVIDIA/cccl/issues/1440
- [ ] https://github.com/NVIDIA/cccl/issues/1536
- [ ] https://github.com/NVIDIA/cccl/issues/1444
- [ ] [FEA]: Add `cuda::ptx::setmaxnreg`
- [ ] [FEA]: Add `cuda::ptx::cp_async_bulk_prefetch`
- [ ] [FEA]: Add `cuda::ptx::cp_async_bulk_prefetch_tensor`
- [ ] [FEA]: Add `cuda::ptx::multimem_*`
- [ ] [FEA]: Add `cuda::ptx::prefetch_tensormap`
- [ ] [FEA]: Add `cuda::ptx::griddepcontrol`
- [ ] [FEA]: Add `cuda::ptx::mbarrier_complete_tx`
- [ ] [FEA]: Add `cuda::ptx::mbarrier_expect_tx`
- [ ] [FEA]: Add `cuda::ptx::stmatrix`
- [ ] [FEA]: Add `cuda::ptx::atom` (cluster scope)
- [ ] [FEA]: Add `cuda::ptx::red` (cluster scope)
- [ ] https://github.com/NVIDIA/cccl/issues/1490

[ahendriksen]

Removing cuda::ptx::mapa again due to bug. See #1442 #1414