Dirty workaround for a buffer overflow crash (with LM3)

Running `Luigi's Mansion 3` results in the following crash at game
startup:

```
$ gdb --args ./bin/yuzu

(gdb) r  # and load luigi's mansion 3 game
...

*** buffer overflow detected ***: /home/user/repos/yuzu-mainline/build/bin/yuzu terminated

Thread 27 "EmuThread" received signal SIGABRT, Aborted.
[Switching to Thread 0x7fffcf7fe700 (LWP 2435)]
__GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
50	../sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) bt
(gdb) bt
 #0  0x00007ffff62ef3eb in __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
 #1  0x00007ffff62ce899 in __GI_abort () at abort.c:79
 #2  0x00007ffff633938e in __libc_message (action=action@entry=(do_abort | do_backtrace), fmt=fmt@entry=0x7ffff6462199 "*** %s ***: %s terminated\n")
    at ../sysdeps/posix/libc_fatal.c:181
 #3  0x00007ffff63dbe57 in __GI___fortify_fail_abort (need_backtrace=need_backtrace@entry=true, msg=msg@entry=0x7ffff6462125 "buffer overflow detected") at fortify_fail.c:33
 #4  0x00007ffff63dbe77 in __GI___fortify_fail (msg=msg@entry=0x7ffff6462125 "buffer overflow detected") at fortify_fail.c:44
 #5  0x00007ffff63da6b6 in __GI___chk_fail () at chk_fail.c:28
 #6  0x0000555555a00639 in memcpy (__len=<optimized out>, __src=<optimized out>, __dest=0x7fffcf7fd2f0) at /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34
 #7  0x0000555555a00639 in Service::Nvidia::Devices::nvhost_ctrl_gpu::GetGpuTime(std::vector<unsigned char, std::allocator<unsigned char> > const&, std::vector<unsigned char, std::allocator<unsigned char> >&) (this=this@entry=0x555557a689e0, input=std::vector of length 16, capacity 16 = {...}, output=std::vector of length 16, capacity 16 = {...})
    at /home/user/repos/yuzu-mainline/src/core/hle/service/nvdrv/devices/nvhost_ctrl_gpu.cpp:197

    at ../sysdeps/posix/libc_fatal.c:181
    at /home/user/repos/yuzu-mainline/src/core/hle/service/nvdrv/devices/nvhost_ctrl_gpu.cpp:197
 #8  0x0000555555a008b6 in Service::Nvidia::Devices::nvhost_ctrl_gpu::ioctl(Service::Nvidia::Devices::nvdevice::Ioctl, std::vector<unsigned char, std::allocator<unsigned char> > const&, std::vector<unsigned char, std::allocator<unsigned char> > const&, std::vector<unsigned char, std::allocator<unsigned char> >&, std::vector<unsigned char, std::allocator<unsigned char> >&, Service::Nvidia::IoctlCtrl&, Service::Nvidia::IoctlVersion)
    (this=0x555557a689e0, command=..., input=std::vector of length 16, capacity 16 = {...}, input2=..., output=std::vector of length 16, capacity 16 = {...}, output2=std::vector of length 0, capacity 0, ctrl=..., version=Service::Nvidia::IoctlVersion::Version1) at /home/user/repos/yuzu-mainline/src/core/hle/service/nvdrv/devices/nvhost_ctrl_gpu.cpp:42
```

I am running Ubuntu 19.10 with `nvidia-driver-440` driver here.

Hopefully, this information will enable you to find the right fix!

src/core/hle/service/nvdrv/devices/nvhost_ctrl_gpu.h

@@ -160,8 +160,9 @@ private:
 
     struct IoctlGetGpuTime {
         u64_le gpu_time;
+        u64_le reserved;
     };
-    static_assert(sizeof(IoctlGetGpuTime) == 8, "IoctlGetGpuTime is incorrect size");
+    static_assert(sizeof(IoctlGetGpuTime) == 16, "IoctlGetGpuTime is incorrect size");
 
     u32 GetCharacteristics(const std::vector<u8>& input, std::vector<u8>& output,
                            std::vector<u8>& output2, IoctlVersion version);