- Minor changes to how hash values are computed for scrambling in the
ZSobolSampler.
- Folded util/bits.h into util/math.h.

No functional changes, save for ZSobolSampler following the other
Sobol Samplers and only warning but not enforcing a pow2 sample
count.

Renamed RandomSampler -> IndependentSampler and thus, now it's
"independent" in scene description files rather than "random" to choose
that one.

Includes various other minor changes that don't affect functionality.

Issue #105.

Apparently Manjaro Linux ships with an unrelated /usr/bin/bin2c.

Issue #105.

With Interval vector elements, Dot() is unaware that each term must be
positive (since it's the square of an interval) and in turn an interval
that spanned zero would occasionally be computed, leading to NaNs in the
following code.

Fixes issue #109.

(Also, small change to hashing in the PaddedSobolSampler.)

Removed support for double-precision Sobol' samples.

Issue #108.

Attempt to fix #107.

When building checkcuda nvcc might not be on the $PATH so instead rely on CMAKE_CUDA_COMPILER.

It's important to randomize the last digit in that case in order to break
correlations between samples returned by calls to Get2D() for different
dimensions...

Bug report and suggested fix via Brennan Shacklett.

Set PBRT_DEBUG_BUILD via generator expression.

Use -march=native with CUDA builds again using a generator expression that
adds '-Xcompiler' for CUDA targets.  (Hopefully works around #87.)
This gives about a 10% perf. benefit with CPU rendering.

Assorted cleanups and improvements in the BilinearPatch code.

Fixes to a few subtle and probably never-encountered bugs (e.g. related to
updating dndu and dndv to account for per-vertex texture coordinates
changing the underlying parameterization...)

(This hopefully will fix the test failures on Ubuntu.)

Assorted variable renamings in SampleSphericalTriangle() and InvertSphericalTriangleSample().

Switch to Floats rather than doubles in InvertSphericalTriangleSample() ->
~5% speedup on the GPU for scenes with triangular light sources.

This is @pierremoreau's suggested fix for issue #100.

I can confirm everything still works on Linux with this fix, but have not
checked the Windows+GPU build.

Issue #104.

Multiple (mostly minor) changes:

- Time is removed from the ray weight computed by the RealisticCamera
  (it's handled by the PixelSensor).
- The "fnumber" and "c" parameters have been removed from the
  PixelSensor's parameters.
- The 683 scale has been removed from both the PixelSensor imaging ratio
  and from SpectrumToPhotometric() (so a no-op, mathematically.)
- Bugfix: Medium implementations keep a separate scalar scale for their
  emission, which is now normalized like regular light sources. This will
  cause image differences.

  diffs for scenes with volumetric emission.