Cycles: add single program debug option for split kernel

Single program generally compiles kernels faster (2-3 times), loads faster,
takes less drive space (2-3 times), and reduces the number of cached kernels.

Cycles: split kernel_shadow_blocked to AO & DL parts

Reduces memory allocation for split kernel.

This allows for faster rendering due to bigger global size,
specially when GPU memory is limited.

Perfromance results:

R9 290 total render time
Before After Change
BMW 4:37 4:34 -1.1 %
Classroom 14:43 14:30 -1.5 %
Fishy Cat 11:20 11:04 -2.4 %
Koro 12:11 12:04 -1.0 %
Pabellon Barcelona 22:01 20:44 -5.8 %
Pabellon Barcelona(*) 15:32 15:09 -2.5 %

(*) without glossy connected to volume

Cycles: Speedup transparent shadows in split kernel

This commit enables record-all transparent shadows rays.

Perfromance results:

R9 290 render time (without synchronization), seconds
Before After Change
BMW 261.5 262.5 +0.4 %
Classroom 869.6 867.3 -0.3 %
Fishy Cat 657.4 639.8 -2.7 %
Koro 1909.8 692.8 -63.7 %
Pabellon Barcelona 1633.3 1238.0 -24.2 %
Pabellon Barcelona(*) 1158.1 903.8 -22.0 %

(*) without glossy connected to volume

Cycles: SSS and Volume rendering in split kernel

Decoupled ray marching is not supported yet.

Transparent shadows are always enabled for volume rendering.

Changes in kernel/bvh and kernel/geom are from Sergey.
This simiplifies code significantly, and prepares it for
record-all transparent shadow function in split kernel.

Cycles: Fix building of CUDA split kernel

Cycles: Fix indentation

Cycles: Fix strict warning about unused variable

Cycles: Calculate size of split state buffer kernel side

By calculating the size of the state buffer in the kernel rather than the host
less code is needed and the size actually reflects the requested features.

Will also be a little faster in some cases because of larger global work size.

Cycles: Fix crash after failed kernel build

Pointers to kernels were uninitialized leading to freeing of random memory
addresses. Another reason it would be good to use smart pointers.

Cycles: Faster building of split kernel

Simple change to make it so that only kernels that have been modified are
rebuilt. Might only be useful during development.

Cycles: Initialize rng_state for split kernel

Because the split kernel can render multiple samples in parallel it is
necessary to have everything initialized before rendering of any samples
begins. The code that normally handles initialization of
`rng_state` (`kernel_path_trace_setup()`) only does so for the first sample,
which was causing artifacts in the split kernel due to uninitialized
`rng_state` for some samples.

Note that because the split kernel can render samples in parallel this
means that the split kernel is incompatible with the LCG.

Cycles: Remove sum_all_radiance kernel

This was only needed for the previous implementation of parallel samples. As
we don't have that any more it can be removed.

Real reason for removal tho is this: `per_sample_output_buffers` was being
calculated too small and artifacts resulted. The tile buffer is already
the correct size and calculating the size for `per_sample_output_buffers`
is a bit difficult with the current layout of the code. As
`per_sample_output_buffers` was only needed for `sum_all_radiance`,
removing that kernel and writing output to the tile buffer directly
fixes the artifacts.

Cycles: Split path initialization into own kernel

This makes it easier to initialize things correctly in the data_init kernel
before they are needed by path tracing.

Cycles: Add names to buffer allocations

This is to help debug and track memory usage for generic buffers. We
have similar for textures already since those require a name, but for
buffers the name is only for debugging proposes.

Cycles: Seperate kernel loading time from render time

Cycles: CUDA implementation of split kernel

Cycles: CPU implementation of split kernel

Cycles: Remove ccl_fetch and SOA

Cycles: Report device maximum allocation and detected global size

Cycles: Workaround for driver hangs

Simple workaround for some issues we've been having with AMD drivers hanging
and rendering systems unresponsive. Unfortunately this makes things a bit
slower, but its better than having to do hard reboots. Will be removed when
drivers have been fixed.

Define CYCLES_DISABLE_DRIVER_WORKAROUNDS to disable for testing purposes.