Build improvements

Cleanup: Remove redundant /Oi flag

Kernel Lifetime Reform Pt. 2

Warn if a new thread is intended to be run on the system CPU core

`/Oi` is included in `/Ox` already

CMake: Enable VS parallel builds for a good reduction in compile times

Passing -pthread to GCC as a flag makes it both link to libpthread, and make C standard library routines reentrant. This makes the additional explicit links unnecessary.

Additionally, on OSX, this is the default behavior, and clang will print a message about it being unused if it's present there.

These are implicitly linked by Xcode, but with this, you can also build it with any other generator, which does not have this behavior.

CoreFoundation is included as a part of Cocoa (which is an umbrella framework), and Cocoa is generally recommended to link against, rather than its individual components (CoreFoundation, Foundation, libobjc, ...).

This both reduces redundancy in add_executable definitions, and makes it easier to link additional libraries. In particular, extra libraries are needed on OSX - see next commit.

This enables the /MP compiler flag, which parallelizes builds of by
distributing compilation of individual object files across workes
processes, instead of being limited to per-project parallelism.

Reduces the time for a full compile from 72 s to 45 s on my machine.

Implemented timers

Core: Fixed a crash and removed some unused variables.

ARM_Disasm only has static methods, so there's no need to have an instance of it.

Timing: Use CoreTiming::GetTicks to keep track of ticks.

This will keep track of idle ticks for us, and fixes some tickcount-related issues

SVC: Fixed SleepThread

It will now properly wait the specified number of nanoseconds and then wake up the thread.

dyncom: Fix UMAAL

These need to be done as a 64-bit operation.

Add check for valid address to call stack

 Threads: Use a dummy idle thread when no other are ready. 

This thread will not actually execute instructions, it will only advance the timing/events and try to yield immediately to the next ready thread, if there aren't any ready threads then it will be rescheduled and start its job again.

GPU: Toggle active framebuffer each frame

Common: Clean up ThreadQueueList

Warn if a new thread is intended to be run on the system CPU core until we implement correct scheduling for such a thread.

dyncom: Fix SMULWB/SMULWT

Wasn't doing proper sign-extension

Replace all the C-style complicated buffer management with a std::deque.
In addition to making the code easier to understand it also adds support
for non-POD IdTypes.

Also clean the rest of the code to follow our code style.