common/logging: Silence no return value warnings

Revert "common/bit_field: Silence sign-conversion warnings"

common/bit_field: Silence sign-conversion warnings

gl_rasterizer: Emulate viewport flipping with ARB_clip_control

core: Make most implicit type conversion warnings errors on MSVC

We can just use numeric_limits instead of relying on wraparound behavior
here.

common_funcs: Remove semicolons from INSERT_PADDING_* macros

Makes code that uses the macros consistent by requiring the lines to be
terminated with a semicolon.

Complete the implementation of RGBA16UI

Implement stub for IApplicationFunctions::QueryApplicationPlayStatisticsByUid

core: Migrate off deprecated mbedtls functions

key_manager: Make use of IOFile in WriteKeyToFile()

shader: Implement FSWZADD and reimplement SHFL

common/hash: Remove unused HashableStruct

xts_archive: Remove redundant std::string constructor

This is unused, so it can be removed. There's better ways of ensuring
zeroed out padding bits, like using zero-initialization, anyhow.

common_funcs: silence sign-conversion warnings in MakeMagic()

We can just call the .data() member of path instead of constructing a
completely new string.

We can trivially resolve these by casting the characters to unsigned
values and then shifting the bits.

video_core: Treat implicit conversions as errors

Move source in windows uploads into its own internal zip

This should prevent path length issues when extracting the build from the installer

ci: Package source with builds

This properly handles unicode-based paths on Windows, while opening a
raw stream doesn't out-of-the-box.

Prevents file creation from potentially failing on Windows PCs that make
use of unicode characters in their save paths (e.g. writing to a user's
AppData folder, where the user has a name with non-ASCII characters).

These functions are marked for deprecation and it's recommended that the
*_ret variants be used instead.

Since the introduction of this library, numerous improvements have been
made. Notably, many of the warnings we would get by simply including the
library header have now been fixed. This makes it much easier to make
conversion warning an error.

These are fairly trivial to resolve and most of the changes entail
using RESULT_UNKNOWN over ResultCode(-1).

We simply need to turn the literal argument to std::accumulate into a
double, rather than an int.

Uncovered a bug within Thread's SetCoreAndAffinityMask() where an
unsigned variable (ideal_core) was being compared against "< 0", which
would always be a false condition.

We can also get rid of an unused function (GetNextProcessorId) which contained a sign
mismatch warning.

Resolves a few trivial sign conversion/mismatch errors.

Will be used to reduce the overall duplication of the same magic value
all over the codebase in following changes.

The constructor was implicitly using signed->unsigned conversions to
produce 0xFFFFFFFF. We can just specify this explicitly with UINT32_MAX.

While we're at it, this also resolves a type truncation warning as well,
given the code was truncating from a 64-bit value to a 32-bit one.