webapi, fake: Navigator.sendBeacon

This is a noop implementation of navigator.sendBeacon. It often shows up in the
logs. I believe that returning "true" to signal successful queuing is correct
as it'll prevent any attempts to fallback. However, I'm less sure that noop'ing
the entire thing is better than just implementing it.

src/browser/webapi/Navigator.zig

@@ -122,6 +122,13 @@ pub fn javaEnabled(_: *const Navigator) bool {
     return false;
 }
 
+/// Noop, signal that the data was successfully queued
+pub fn sendBeacon(_: *const Navigator, url: js.Value, data: ?js.Value) bool {
+    _ = url;
+    _ = data;
+    return true;
+}
+
 pub fn getPlugins(self: *Navigator) *PluginArray {
     return &self._plugins;
 }
@@ -239,6 +246,7 @@ pub const JsApi = struct {
     pub const globalPrivacyControl = bridge.accessor(Navigator.getGlobalPrivacyControl, null, .{});
 
     pub const javaEnabled = bridge.function(Navigator.javaEnabled, .{});
+    pub const sendBeacon = bridge.function(Navigator.sendBeacon, .{ .exposed = .window, .noop = true });
     pub const permissions = bridge.accessor(Navigator.getPermissions, null, .{});
     pub const storage = bridge.accessor(Navigator.getStorage, null, .{});
     pub const userAgentData = bridge.accessor(Navigator.getUserAgentData, null, .{});