fix pystack exhaust during resume_process_key.

Instead of handling resumed key events in a deep stack, buffer them
until the next main loop iteration. New resume events that may be emitted
during handling of old resumes are prepended to that buffer, i.e. take
precedence over events that happen deeper into the buffer/event stack.
Logical replay order is thus preserved.

kmk/modules/layers.py

@@ -1,6 +1,6 @@
 '''One layer isn't enough. Adds keys to get to more of them'''
 from kmk.keys import KC, make_argumented_key
-from kmk.modules.holdtap import ActivationType, HoldTap, HoldTapKeyMeta
+from kmk.modules.holdtap import HoldTap, HoldTapKeyMeta
 from kmk.utils import Debug
 
 debug = Debug(__name__)
@@ -73,20 +73,6 @@ class Layers(HoldTap):
             on_release=self.ht_released,
         )
 
-    def process_key(self, keyboard, key, is_pressed, int_coord):
-        current_key = super().process_key(keyboard, key, is_pressed, int_coord)
-
-        for key, state in self.key_states.items():
-            if key == current_key:
-                continue
-
-            # on interrupt: key must be translated here, because it was asigned
-            # before the layer shift happend.
-            if state.activated == ActivationType.INTERRUPTED:
-                current_key = keyboard._find_key_in_map(int_coord)
-
-        return current_key
-
     def _df_pressed(self, key, keyboard, *args, **kwargs):
         '''
         Switches the default layer