Let's yuzu_cmd use a network interface. Also adds it to the default ini.

* better_description

* Revert "better_description"

This reverts commit 3a152a6ba6f2d6e02530b69a8194e4db302c8acd.

* better_nvdec_wording

* best_performance

* update_word_stuff

* another_update

yuzu qt: Remove global system instances

vic: Use the minimum of surface/frame dimensions when writing the final frame to the GPU

input_common/sdl: Fix joystick range

main: Add option to reset window size to 900p

externals: Update dynarmic to cce7e4ee

Update configure_tas.ui

h264: Use max allowed max_num_ref_frames when using CPU decoding

Port citra-emu/citra#5811: "Build system: remove references to travis"

Co-Authored-By: Vitor K <vitor-kiguchi@hotmail.com>

This moves all GenerateYuzuPath calls outside of the platofrm-specific #ifdefs, replacing them with assignments to paths.

Fix a few warnings

- configure_input_player_widget.cpp: always better to use `const auto &`
  whenever possible
- profiler.cpp: `ev->pos()` is deprecated, replace with
  `ev->position()`, which returns floats, thus the addition of
  `.toPoint()` (same as what's happening in `pos()`)
- game_list.cpp: `QString::SplitBehavior` is deprecate, use `Qt::`
  namespace instead

ci: Use ubuntu-latest vmImage where applicable

Not specifying the vmImage defaults to ubuntu-16.04, which will be deprecated soon and is experiencing brownouts.

applets/web: Fallback to loader to get the manual romfs if none is found

Addresses possible buffer overflow behavior.

vic: Allow surface to be higher than frame

Touhou Genso Wanderer Lotus Labyrinth R decodes 1920x1080 videos into 1920x1088 surface.
Only allow mismatch for height, since larger width would result in increasingly offset rows and somewhat defeat entire purpose of this check.

vic: Implement RGBX8 video frame format

Remove the geometry property again(7045) after 7090 re-added.

This is a work around to avoid buffer overflow errors until multi channel/multi stream decoding is supported.

service: Reduce header include overhead

gpu: Migrate implementation to the cpp file