Setting an object name for GPUCommandStreamWidget allows for saving the
graphics debugger's state (if it's show, position, etc). This state is
then restored when restarting the application.

Stubbed CreateMemoryBlock

Using Berkeley sockets, and Winsock2.2 on Windows.
So far ftpony creates the socket and accepts incoming connections

SOC_U: Renamed functions to maintain consistency

Also prevents possible scope errors / conflicts with the actual Berkeley socket functions

SOCU: Close all the opened sockets when cleaning up SOCU

They will be stored in /extsavedata/SDMC and /extsavedata/NAND respectively.
Also redirect some APT_A functions to their APT_U equivalents.
Implemented the gamecoin.dat file in SharedExtSaveData in the PTM module.
Implemented formatting the savegame.
Retake a previous savegame if it exists instead of reporting them as not formatted every time a game is loaded.

They were erroneously labeled as SMLAL.

Execute the division-by-W for perspective-correct interpolation of
values in the clipper, moving them out of the rasterization inner loop.

Replace the loop-based texture address swizzling code by a bit-twiddling
implementation, providing a very small speed up. Also simplify
addressing code.

Move the computation of some values out of loops so that they're not
constantly recalculated even when they don't change.

Use a new buffer management scheme in the clipper that allows using a
bounded minimal amount of buffer space. Even though it copies more data
it is still slightly faster likely due to using less cache.

The triangle clipper was allocating its temporary input, output and work
buffers using a std::vector. Since this is a hot path, it's desirable to
use stack allocation instead.

Unused OutputVertex attributes were being left un-initialized. The
leftover garbage sometimes decoded as floating-point denormalized
values, causing fallbacks to microcode and massive slowdowns in the rest
of the rasterization pipeline even though the results were unused. By
zeroing the structure we ensure these attributes only contain harmless
zeros.

This handle manager more closely mirrors the behaviour of the CTR-OS
one. In addition object ref-counts and support for DuplicateHandle have
been added.

Note that support for DuplicateHandle is still experimental, since parts
of the kernel still use Handles internally, which will likely cause
troubles if two different handles to the same object are used to e.g.
wait on a synchronization primitive.

Initially reported by xdec.