Change to allow cccl/c/parallel/unique_by_key.h to compile by C compiler

[oleksandr-pavlyk]

1. Remove noexcept
2. Use uint64_t instead of unsigned long long (no required to allow compilation by C compiler, but adopted throughout to one day support compilation on Windows)
3. Add #include <unique_by_key.h> to test_headers.c test file

Description

closes #4258

Checklist

• New or existing tests cover these changes.
• The documentation is up to date with these changes.

[oleksandr-pavlyk]

While at it, is there a reason by num_items argument of cccl_unique_by_key is placed after cccl_operator_t comparison_op and not before as, for example, is done in cccl_merge_sort?

[NaderAlAwar]

While at it, is there a reason by num_items argument of cccl_unique_by_key is placed after cccl_operator_t comparison_op and not before as, for example, is done in cccl_merge_sort?

The reason I did it was to follow the order of the arguments in DispatchUniqueByKey::Dispatch()

cccl/cub/cub/device/dispatch/dispatch_unique_by_key.cuh

Lines 487 to 500 in 5d22600

	CUB_RUNTIME_FUNCTION _CCCL_FORCEINLINE static cudaError_t Dispatch(
	void* d_temp_storage,
	size_t& temp_storage_bytes,
	KeyInputIteratorT d_keys_in,
	ValueInputIteratorT d_values_in,
	KeyOutputIteratorT d_keys_out,
	ValueOutputIteratorT d_values_out,
	NumSelectedIteratorT d_num_selected_out,
	EqualityOpT equality_op,
	OffsetT num_items,
	cudaStream_t stream,
	KernelSource kernel_source = {},
	KernelLauncherFactory launcher_factory = {},
	MaxPolicyT max_policy = {})

However looking at the API I can see now that num_items comes before, so I think we should maybe switch it here too. I can open another PR but you can also feel free to do it here.

[NaderAlAwar]

LGTM, thanks @oleksandr-pavlyk

[github-actions]

🟩 CI finished in 1h 11m: Pass: 100%/3 | Total: 1h 30m | Avg: 30m 09s | Max: 1h 10m | Hits: 95%/328

• 🟩 cccl_c_parallel: Pass: 100%/2 | Total: 20m 09s | Avg: 10m 04s | Max: 17m 25s | Hits: 95%/328

  🟩 cpu
    🟩 amd64              Pass: 100%/2   | Total: 20m 09s | Avg: 10m 04s | Max: 17m 25s | Hits:  95%/328   
  🟩 ctk
    🟩 12.8               Pass: 100%/2   | Total: 20m 09s | Avg: 10m 04s | Max: 17m 25s | Hits:  95%/328   
  🟩 cudacxx
    🟩 nvcc12.8           Pass: 100%/2   | Total: 20m 09s | Avg: 10m 04s | Max: 17m 25s | Hits:  95%/328   
  🟩 cudacxx_family
    🟩 nvcc               Pass: 100%/2   | Total: 20m 09s | Avg: 10m 04s | Max: 17m 25s | Hits:  95%/328   
  🟩 cxx
    🟩 GCC13              Pass: 100%/2   | Total: 20m 09s | Avg: 10m 04s | Max: 17m 25s | Hits:  95%/328   
  🟩 cxx_family
    🟩 GCC                Pass: 100%/2   | Total: 20m 09s | Avg: 10m 04s | Max: 17m 25s | Hits:  95%/328   
  🟩 gpu
    🟩 rtx2080            Pass: 100%/2   | Total: 20m 09s | Avg: 10m 04s | Max: 17m 25s | Hits:  95%/328   
  🟩 jobs
    🟩 Build              Pass: 100%/1   | Total:  2m 44s | Avg:  2m 44s | Max:  2m 44s | Hits:  92%/164   
    🟩 Test               Pass: 100%/1   | Total: 17m 25s | Avg: 17m 25s | Max: 17m 25s | Hits:  98%/164   
  

• 🟩 python: Pass: 100%/1 | Total: 1h 10m | Avg: 1h 10m | Max: 1h 10m

  🟩 cpu
    🟩 amd64              Pass: 100%/1   | Total:  1h 10m | Avg:  1h 10m | Max:  1h 10m
  🟩 ctk
    🟩 12.8               Pass: 100%/1   | Total:  1h 10m | Avg:  1h 10m | Max:  1h 10m
  🟩 cudacxx
    🟩 nvcc12.8           Pass: 100%/1   | Total:  1h 10m | Avg:  1h 10m | Max:  1h 10m
  🟩 cudacxx_family
    🟩 nvcc               Pass: 100%/1   | Total:  1h 10m | Avg:  1h 10m | Max:  1h 10m
  🟩 cxx
    🟩 GCC13              Pass: 100%/1   | Total:  1h 10m | Avg:  1h 10m | Max:  1h 10m
  🟩 cxx_family
    🟩 GCC                Pass: 100%/1   | Total:  1h 10m | Avg:  1h 10m | Max:  1h 10m
  🟩 gpu
    🟩 rtx2080            Pass: 100%/1   | Total:  1h 10m | Avg:  1h 10m | Max:  1h 10m
  🟩 jobs
    🟩 Test               Pass: 100%/1   | Total:  1h 10m | Avg:  1h 10m | Max:  1h 10m
  

👃 Inspect Changes

Modifications in project?

	Project
	CCCL Infrastructure
	libcu++
	CUB
	Thrust
	CUDA Experimental
	stdpar
	python
+/-	CCCL C Parallel Library
	Catch2Helper

Modifications in project or dependencies?

	Project
	CCCL Infrastructure
	libcu++
	CUB
	Thrust
	CUDA Experimental
	stdpar
+/-	python
+/-	CCCL C Parallel Library
	Catch2Helper

🏃‍ Runner counts (total jobs: 3)

#	Runner
2	linux-amd64-gpu-rtx2080-latest-1
1	linux-amd64-cpu16