Use Remote Processor Framework (remoteproc) of Linux directly instead of OpenCL

[henrix]

OpenCL is not suitable for DSP applications due to unpredictable latencies of task queuing.
The performance of the library can be heavily increased by getting rid of OpenCL and using remoteproc directly instead.

[JayKickliter]

remoteproc is just a subsystem for controlling non-host processors, plus some messaging (rpmsg). Without OpenCL, you're left to writing+compiling your remote firmware and defining the API AoT. But, FWIW, I do agree that OpenCL doesn't seem like the best abstraction for the C66xs.