Add test_reduce_api.py::test_reduce_struct_type_minmax

[oleksandr-pavlyk]

This adds an example of using struct that holds minimum and maximum values to compute smallest and largest values of an array using a single call cub::DeviceReduce

It also demonstrates a use of TransformedIterator to map input array values as v -> MinMax(v,v) to make reduction binary operator commutative, to render data-parallel algorithm applicable. Since this is done in registers, each element of input array is only accessed once.

Docstring of gpu_struct is modified to reference newly added example.

Description

closes gh-3937

Checklist

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

[github-actions]

🟥 CI finished in 7m 40s: Pass: 0%/1 | Total: 7m 40s | Avg: 7m 40s | Max: 7m 40s

• 🟥 python: Pass: 0%/1 | Total: 7m 40s | Avg: 7m 40s | Max: 7m 40s

  🟥 cpu
    🟥 amd64              Pass:   0%/1   | Total:  7m 40s | Avg:  7m 40s | Max:  7m 40s
  🟥 ctk
    🟥 12.8               Pass:   0%/1   | Total:  7m 40s | Avg:  7m 40s | Max:  7m 40s
  🟥 cudacxx
    🟥 nvcc12.8           Pass:   0%/1   | Total:  7m 40s | Avg:  7m 40s | Max:  7m 40s
  🟥 cudacxx_family
    🟥 nvcc               Pass:   0%/1   | Total:  7m 40s | Avg:  7m 40s | Max:  7m 40s
  🟥 cxx
    🟥 GCC13              Pass:   0%/1   | Total:  7m 40s | Avg:  7m 40s | Max:  7m 40s
  🟥 cxx_family
    🟥 GCC                Pass:   0%/1   | Total:  7m 40s | Avg:  7m 40s | Max:  7m 40s
  🟥 gpu
    🟥 rtx2080            Pass:   0%/1   | Total:  7m 40s | Avg:  7m 40s | Max:  7m 40s
  🟥 jobs
    🟥 Test               Pass:   0%/1   | Total:  7m 40s | Avg:  7m 40s | Max:  7m 40s
  

👃 Inspect Changes

Modifications in project?

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

Modifications in project or dependencies?

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

🏃‍ Runner counts (total jobs: 1)

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

[github-actions]

🟩 CI finished in 50m 06s: Pass: 100%/1 | Total: 50m 06s | Avg: 50m 06s | Max: 50m 06s

• 🟩 python: Pass: 100%/1 | Total: 50m 06s | Avg: 50m 06s | Max: 50m 06s

  🟩 cpu
    🟩 amd64              Pass: 100%/1   | Total: 50m 06s | Avg: 50m 06s | Max: 50m 06s
  🟩 ctk
    🟩 12.8               Pass: 100%/1   | Total: 50m 06s | Avg: 50m 06s | Max: 50m 06s
  🟩 cudacxx
    🟩 nvcc12.8           Pass: 100%/1   | Total: 50m 06s | Avg: 50m 06s | Max: 50m 06s
  🟩 cudacxx_family
    🟩 nvcc               Pass: 100%/1   | Total: 50m 06s | Avg: 50m 06s | Max: 50m 06s
  🟩 cxx
    🟩 GCC13              Pass: 100%/1   | Total: 50m 06s | Avg: 50m 06s | Max: 50m 06s
  🟩 cxx_family
    🟩 GCC                Pass: 100%/1   | Total: 50m 06s | Avg: 50m 06s | Max: 50m 06s
  🟩 gpu
    🟩 rtx2080            Pass: 100%/1   | Total: 50m 06s | Avg: 50m 06s | Max: 50m 06s
  🟩 jobs
    🟩 Test               Pass: 100%/1   | Total: 50m 06s | Avg: 50m 06s | Max: 50m 06s
  

👃 Inspect Changes

Modifications in project?

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

Modifications in project or dependencies?

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

🏃‍ Runner counts (total jobs: 1)

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

[github-actions]

🟩 CI finished in 50m 53s: Pass: 100%/1 | Total: 50m 53s | Avg: 50m 53s | Max: 50m 53s

• 🟩 python: Pass: 100%/1 | Total: 50m 53s | Avg: 50m 53s | Max: 50m 53s

  🟩 cpu
    🟩 amd64              Pass: 100%/1   | Total: 50m 53s | Avg: 50m 53s | Max: 50m 53s
  🟩 ctk
    🟩 12.8               Pass: 100%/1   | Total: 50m 53s | Avg: 50m 53s | Max: 50m 53s
  🟩 cudacxx
    🟩 nvcc12.8           Pass: 100%/1   | Total: 50m 53s | Avg: 50m 53s | Max: 50m 53s
  🟩 cudacxx_family
    🟩 nvcc               Pass: 100%/1   | Total: 50m 53s | Avg: 50m 53s | Max: 50m 53s
  🟩 cxx
    🟩 GCC13              Pass: 100%/1   | Total: 50m 53s | Avg: 50m 53s | Max: 50m 53s
  🟩 cxx_family
    🟩 GCC                Pass: 100%/1   | Total: 50m 53s | Avg: 50m 53s | Max: 50m 53s
  🟩 gpu
    🟩 rtx2080            Pass: 100%/1   | Total: 50m 53s | Avg: 50m 53s | Max: 50m 53s
  🟩 jobs
    🟩 Test               Pass: 100%/1   | Total: 50m 53s | Avg: 50m 53s | Max: 50m 53s
  

👃 Inspect Changes

Modifications in project?

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

Modifications in project or dependencies?

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

🏃‍ Runner counts (total jobs: 1)

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