* Fix GLFW runtime platform detection

* Add libwayland-dev package to workflows

* Add libxkbcommon-dev package to workflows

* Add libwayland-bin package to codeql.yml

* Add libwayland-client0 and libwayland-cursor0 packages to codeql.yml

* Use libwayland-dev in codeql.yml

* Add libxkbcommon-dev to codeql.yml

* Remove libwayland-bin from linux.yml and linux_examples.yml (libwayland-dev includes it as a dependency)

---------

Co-authored-by: Ray <raysan5@gmail.com>

Odin's binding were bumped to 5.0 back in November 2023

Less matrix operations doing.

I did some testing and this seems to work okay, but it *is* different to
GLFW. GLFW seems to lose "IsCursorOnScreen" for one frame upon pressing
alt+tab over the window, without a different window covering it. SDL
seems to have a more reasonable behavior for this case. Either way it's
an edge case and I don't know the exact way to fix, likely requires
internal GLFW changes.

* cmake: For SDL, disable rlgl_standalone

* cmake: Only build rlgl_standalone when on Desktop, and not shared libs

* Screen/world-space related functions rename

* Update raylib_api.* by CI

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

* CMake: Remove USE_WAYLAND option

* Consistency fix

* Fix oversight

The fingerId from SDL was used as an index into the CORE.Input.Touch
arrays, but it's an opaque / arbitrary int64, way bigger
than MAX_TOUCH_POINTS, so the first non-simulated touch event would segfault.

* Update GLFW to 3.4 (draft)

* Add _glfwConnectNull() function to rglfw.c

* Update rglfw.c

* Update Makefile

* Makefile: Replace USE_WAYLAND_DISPLAY with DISABLE_WAYLAND

* Revert "Makefile: Replace USE_WAYLAND_DISPLAY with DISABLE_WAYLAND"

This reverts commit 9e79abde78.

* GlfwImport.cmake: Replace GLFW_USE_WAYLAND with GLFW_BUILD_WAYLAND

* Update rglfw.c

* Output platform selected by GLFW to TRACELOG

* GLFW has removed Mir support