gl_device: Enable NV_vertex_buffer_unified_memory on Turing devices

maxwell_to_gl/vk: Reorder vertex formats

Macro code is just uploaded sequentially from a starting address, however that does not mean the entry point for the macro is at that address. This PR adds preliminary support for executing macros in the middle of our cached code.

hotkeys: Fix issues caused when changing the fullscreen hotkey

renderer_vulkan: Update validation layer name and test before enabling

Remake Kernel Scheduling, CPU Management & Boot Management (Prometheus)

ldr: Cleanup NRO & NRR structs

ldr: Update NRR/NRO structs

grc: Update function table

lbl: Update function table

ldn: Update function table

mig: Update function table

mm: Update function table

ncm: Update function table

nfc: Update function table

friend: Update function table

Add 32 bit component sizes to (un)signed scaled formats and group (un)signed normalized, scaled, and integer formats together.