[FEA]: Improve performance of thrust::all_of.

[gonidelis]

Is this a duplicate?

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

Area

Thrust

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

Addresses and provides a roadmap for #720. The current idea is that we explore whether a new cub::DeviceFind::FindIf implementation could be used for a twofold cause: a) port thrust::find to CUB and b) use a common CUB implementation for thrust::all_of that utilizes early exit.

Describe the solution you'd like

Tasks

• Build and benchmark a prototype CUB cub::DeviceFind::FindIf kernel #2114
• Benchmark and optimize cub::DeviceFind::FindIf against thrust::find_if, thrust::count_if. #2115
• Port thrust::all_of to cub::DeviceFind::FindIf. #2116
• Port thrust::find to cub::DeviceFind::FindIf. #2117

Describe alternatives you've considered

No response

Additional context

No response