feat(draw): support transforming widgets and improfe sw transform

For details see: https://docs.lvgl.io/master/overview/style.html

lv_conf_template.h

@@ -120,6 +120,24 @@
     #define LV_CIRCLE_CACHE_SIZE 4
 #endif /*LV_DRAW_COMPLEX*/
 
+/**
+ * "Simple layers" are used when a widget has `style_opa < 255` to buffer the widget into a layer
+ * and blend it as an image with the given opacity.
+ * Note that `bg_opa`, `text_opa` etc don't require buffering into layer)
+ * The widget can be buffered in smaller chunks to avoid using large buffers.
+ * `draw_area` (`lv_area_t` meaning the area to draw and `px_size` (size of a pixel in bytes)
+ * can be used the set the buffer size adaptively.
+ *
+ * - LV_LAYER_SIMPLE_BUF_SIZE: [bytes] the optimal target buffer size. LVGL will try to allocate it
+ * - LV_LAYER_SIMPLE_FALLBACK_BUF_SIZE: [bytes]  used if `LV_LAYER_SIMPLE_BUF_SIZE` couldn't be allocated.
+ *
+ * Both buffer sizes are in bytes.
+ * "Transformed layers" (where transform_angle/zoom properties are used) use larger buffers
+ * and can't be drawn in chunks. So these settings affects only widgets with opacity.
+ */
+#define LV_LAYER_SIMPLE_BUF_SIZE  		   (24 * 1024)
+#define LV_LAYER_SIMPLE_FALLBACK_BUF_SIZE  LV_MAX(lv_area_get_width(&draw_area) * px_size, 2048)
+
 /*Default image cache size. Image caching keeps the images opened.
  *If only the built-in image formats are used there is no real advantage of caching. (I.e. if no new image decoder is added)
  *With complex image decoders (e.g. PNG or JPG) caching can save the continuous open/decode of images.