Add WgpuConfiguration::desired_maximum_frame_latency (#3874)

Setting `desired_maximum_frame_latency` to a low value should
theoretically lead to lower latency in winit apps using `egui-wgpu`
(e.g. in `eframe` with `wgpu` backend).

* Replaces https://github.com/emilk/egui/pull/3714
* See also https://github.com/gfx-rs/wgpu/pull/4899

----

It seems like `desired_maximum_frame_latency` has no effect on my Mac. I
lowered my monitor refresh-rate to 30Hz to test, and can see no
difference between `desired_maximum_frame_latency` of `0` or `3`.

Before when experimenting with changing the global `DESIRED_NUM_FRAMES`
in `wgpu` I saw a huge difference, so I wonder what has changed.

I verified that `set_maximum_drawable_count` is being called with either
`1` or `2`, but I perceive no difference between the two.

crates/egui-wgpu/src/lib.rs

@@ -228,6 +228,15 @@ pub struct WgpuConfiguration {
     /// Present mode used for the primary surface.
     pub present_mode: wgpu::PresentMode,
 
+    /// Desired maximum number of frames that the presentation engine should queue in advance.
+    ///
+    /// Use `1` for low-latency, and `2` for high-throughput.
+    ///
+    /// See [`wgpu::SurfaceConfiguration::desired_maximum_frame_latency`] for details.
+    ///
+    /// `None` = `wgpu` default.
+    pub desired_maximum_frame_latency: Option<u32>,
+
     /// Power preference for the adapter.
     pub power_preference: wgpu::PowerPreference,
 
@@ -237,10 +246,22 @@ pub struct WgpuConfiguration {
 
 impl std::fmt::Debug for WgpuConfiguration {
     fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        let Self {
+            supported_backends,
+            device_descriptor: _,
+            present_mode,
+            desired_maximum_frame_latency,
+            power_preference,
+            on_surface_error: _,
+        } = self;
         f.debug_struct("WgpuConfiguration")
-            .field("supported_backends", &self.supported_backends)
-            .field("present_mode", &self.present_mode)
-            .field("power_preference", &self.power_preference)
+            .field("supported_backends", &supported_backends)
+            .field("present_mode", &present_mode)
+            .field(
+                "desired_maximum_frame_latency",
+                &desired_maximum_frame_latency,
+            )
+            .field("power_preference", &power_preference)
             .finish_non_exhaustive()
     }
 }
@@ -274,6 +295,8 @@ impl Default for WgpuConfiguration {
 
             present_mode: wgpu::PresentMode::AutoVsync,
 
+            desired_maximum_frame_latency: None,
+
             power_preference: wgpu::util::power_preference_from_env()
                 .unwrap_or(wgpu::PowerPreference::HighPerformance),