Avoid 8-bit timer overflows in debounce algorithms (#12240)

* Add fast_timer_t that is 16-bit or 32-bit based on architecture

A 16-bit timer will overflow sooner but be faster to compare on AVR.

* Avoid 8-bit timer overflows in debounce algorithms

Count down remaining elapsed time instead of trying to do 8-bit timer
comparisons.

Add a "none" implementation that is automatically used if DEBOUNCE is
0 otherwise it will break the _pk/_pr count down.

* Avoid unnecessary polling of the entire matrix in sym_eager_pk

The matrix only needs to be updated when a debounce timer expires.

* Avoid unnecessary polling of the entire matrix in sym_eager_pr

The matrix only needs to be updated when a debounce timer expires.

The use of the "needed_update" variable is trying to do what
"matrix_need_update" was added to fix but didn't work because it only
applied when all keys finished debouncing.

* Fix sym_defer_g timing inconsistency compared to other debounce algorithms

DEBOUNCE=5 should process the key after 5ms, not 6ms

* Add debounce tests

tmk_core/common/avr/_timer.h

@@ -0,0 +1,19 @@
+/* Copyright 2021 Simon Arlott
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+#pragma once
+
+// The platform is 8-bit, so prefer 16-bit timers to reduce code size
+#define FAST_TIMER_T_SIZE 16