If a block has assigned a XfbOffset it is assumed that it would
inherit the current global XfbBuffer. This commit fixes two use cases:

1) Getting the members of a Block with a XfbOffset to be assigned an
   offset, as explained on GLSL 4.60 spec, section "4.4.2 Output
   Layout Qualifiers", subsection "Transform Feedback Layout
   Qualifiers".

2) Compute properly an error on overlapping ranges if a block is
   assigned a XfbOffset and one of it members is assigned a explicit
   one. This gets working because when the members of a block get
   assigned a Offset/Buffer at fixBlockXfbOffsets, then the block is
   deassigned the Offsets, so ranges are computed only with the block
   members.

BTW, this is already done when redeclaring block builtins.

Fixes #1535

Apinheiro/xfb struct assignment

Add DeadBranchElim before MergeReturn in legalization.

MergeReturn now requires this or could throw an error.

Rename PCH macro to glslang_pch (to avoid name collision) and update to latest spirv-tools

Use precompiled headers for some glslang projects (MSVC-only)

Add names for composite spec constants in SPIR-V

Consider the following code:

    layout(constant_id=0) const int Y = 1;
    layout(constant_id=1) const int Z = 2;
    layout(constant_id=3) const int X = Y + Z;

Previously, it would produce SPIR-V decorations like this:

    Decorate 21(Y) SpecId 1
    Decorate 22 SpecId 3
    Decorate 33(Z) SpecId 0

This seems inaccurate, since the spec constant `X` that is dependent on
the two others did not get a name in the SPIR-V decorations. This behavior
may potentially negatively affect shader introspection capabilities.

This change alters the behavior to always add a name, which results in the code
above producing the following decorations:

    Decorate 21(Y) SpecId 1
    Decorate 22(X) SpecId 3
    Decorate 33(Z) SpecId 0

Add support for nv_ray_tracing final 

If the out variable is a struct type, with a xfb_offset explicitly
assigned, the members need to get their xfb_offset assigned. This is
specially relevant, as we cannot use layout qualifiers on struct
members.

BUILD.gn: remove reference to non-existent header

Fix MSVC warning C4065: 'default' but no 'case' labels

This would happen in SpvPostProcess when we don't enable AMD extensions.

Fix 8-bit storage nearly always using the UniformAndStorageBuffer8BitAccess capability.

Allow GL_NV_mesh_shader in fragment shaders for perprimitiveNV

UniformAndStorageBuffer8BitAccess capability.

When using the 8-bit storage extension it basically always used the
`UniformAndStorageBuffer8BitAccess` capability, even in cases where it
wasn't required. For instance if we are targeting Vulkan 1.1 (SPIR-V 1.3
or higher), and we are only using 8-bit types in an SSBO, we only need
the `StorageBuffer8BitAccess` capability.

I fixed this by enabling storage buffer use in Vulkan 1.1 / SPIR-V 1.3
or higher, and then changing the logic to match.

I also added some tests that will output different capabilities when run
on Vulkan 1.0 and 1.1, thus they are added twice to the test list (one
for each version).

Fixes #1539

- Emit relevant capability/extension for use of perprimitiveNV in fragment shader
- Remove redundant checks for mesh shader qualifiers in glslang.y
- Add profile version check for use of extension GL_NV_mesh_shader
- Add a new gtest for use of perprimitiveNV in fragment shader

Fixup unused parameter warnings

This CL removes the current parameters which are unused in order to
fixup the issued clang warnings.

BUILD.gn: remove the static_library target.

This had no sources and depended on a source_set which is an invalid use
of GN because the source_set gets propagated to dependents of the
static_library instead.

Fix a parse issue of imageAtomicCompSwap()

The max count of arguments is 5 when image2DMS/image2DMSArray is
specified.

Fix flags.nonprivate to be true for workgroup memory, which is implicitly workgroupcoherent/nonprivate

Fix flags.nonprivate to be true for workgroup memory, which is implicitly workgroupcoherent/nonprivate

Allow redeclaration of builtin interface blocks in mesh shader

Fix inferred output primitive for GL_NV_geometry_shader_passthrough