Desktop defaults to highp for samplers, but this should not apply to the built-in
functions, so make it appy only to user declarations.

Preprocessor: Evaluate INT_MIN / -1 to 0.

HLSL: Partially flatten hierarchies, instead of all or nothing.

  This division is undefined behaviour which raises SIGFPE on x86.
  Most C++ preprocessors evaluate this silently to 0.

GLSL: Don't try to swizzle void.

Also, rationalized this to generally make it safer and more readable.
It could use a more modern approach, at some point...

Fixes #1088.

Fixes #1092.  Allows arrays of opaques to keep arrayness, unless
needed by uniform array flattening.
Can handle assignments of mixed amounts of flattening.

HLSL: nonfunctional: add helper access methods to TAttributeMap

Set cmake-policy CMP0048 to NEW

HLSL: nonfunctional: rename setId -> switchId, add comment

Add getAsLoopNode() method for TIntermLoop

There was some code replication around getting string and integer
values out of an attribute map.  This adds new methods to the
TAttributeMap class to encapsulate some accessor details.

follow local conventions of spaces

Method rename, add comment about its intended use.  No
test diffs.

Fixes #1091.

Only track variables in the global scope

The semantics differ between GLSL/HLSL and SPIR-V.
Translate between these.

HLSL: split textures used for both shadow and non-shadow modes

A single texture can statically appear in code mixed with a shadow sampler
and a non-shadow sampler.  This would be create invalid SPIR-V, unless
one of them is provably dead.

The previous detection of this happened before DCE, so some shaders would
trigger the error even though they wouldn't after DCE.  To handle that
case, this PR splits the texture into two: one with each mode.  It sets
"needsLegalization" (if that happens for any texture) to warn that this shader
will need post-compilation legalization.

If the texture is only used with one of the two modes, behavior is as it
was before.

HLSL: Fix #954: Track/access subsets of flattened multi-level aggrega…

Works in conjuction with d1be7545 to represent and modify a partially
dereferenced multi-level flattened aggregate.

HLSL: Fix crash when flattening both side of assignement simultaneously.