Organize the engine/ directory, director's cut

Cleaned up `engine/routines`, in favor of moving files into more
appropriate directories. predef-related routines are now in top-level
`engine`.
`rtc/delete_save_change_clock.asm` has been split into both
`menus/delete_save.asm` and `rtc/reset_password.asm`.

Made a new subdirectory:
* engine/math: Contains all generic math-related routines.

engine/math/math.asm

@@ -0,0 +1,196 @@
+_Multiply:: ; 66de
+
+; hMultiplier is one byte.
+	ld a, 8
+	ld b, a
+
+	xor a
+	ld [hProduct], a
+	ld [hMathBuffer + 1], a
+	ld [hMathBuffer + 2], a
+	ld [hMathBuffer + 3], a
+	ld [hMathBuffer + 4], a
+
+
+.loop
+	ld a, [hMultiplier]
+	srl a
+	ld [hMultiplier], a
+	jr nc, .next
+
+	ld a, [hMathBuffer + 4]
+	ld c, a
+	ld a, [hMultiplicand + 2]
+	add c
+	ld [hMathBuffer + 4], a
+
+	ld a, [hMathBuffer + 3]
+	ld c, a
+	ld a, [hMultiplicand + 1]
+	adc c
+	ld [hMathBuffer + 3], a
+
+	ld a, [hMathBuffer + 2]
+	ld c, a
+	ld a, [hMultiplicand + 0]
+	adc c
+	ld [hMathBuffer + 2], a
+
+	ld a, [hMathBuffer + 1]
+	ld c, a
+	ld a, [hProduct]
+	adc c
+	ld [hMathBuffer + 1], a
+
+.next
+	dec b
+	jr z, .done
+
+
+; hMultiplicand <<= 1
+
+	ld a, [hMultiplicand + 2]
+	add a
+	ld [hMultiplicand + 2], a
+
+	ld a, [hMultiplicand + 1]
+	rla
+	ld [hMultiplicand + 1], a
+
+	ld a, [hMultiplicand + 0]
+	rla
+	ld [hMultiplicand + 0], a
+
+	ld a, [hProduct]
+	rla
+	ld [hProduct], a
+
+	jr .loop
+
+
+.done
+	ld a, [hMathBuffer + 4]
+	ld [hProduct + 3], a
+
+	ld a, [hMathBuffer + 3]
+	ld [hProduct + 2], a
+
+	ld a, [hMathBuffer + 2]
+	ld [hProduct + 1], a
+
+	ld a, [hMathBuffer + 1]
+	ld [hProduct + 0], a
+
+	ret
+; 673e
+
+
+_Divide:: ; 673e
+	xor a
+	ld [hMathBuffer + 0], a
+	ld [hMathBuffer + 1], a
+	ld [hMathBuffer + 2], a
+	ld [hMathBuffer + 3], a
+	ld [hMathBuffer + 4], a
+
+	ld a, 9
+	ld e, a
+
+.loop
+	ld a, [hMathBuffer + 0]
+	ld c, a
+	ld a, [hDividend + 1]
+	sub c
+	ld d, a
+
+	ld a, [hDivisor]
+	ld c, a
+	ld a, [hDividend + 0]
+	sbc c
+	jr c, .next
+
+	ld [hDividend + 0], a
+
+	ld a, d
+	ld [hDividend + 1], a
+
+	ld a, [hMathBuffer + 4]
+	inc a
+	ld [hMathBuffer + 4], a
+
+	jr .loop
+
+.next
+	ld a, b
+	cp 1
+	jr z, .done
+
+	ld a, [hMathBuffer + 4]
+	add a
+	ld [hMathBuffer + 4], a
+
+	ld a, [hMathBuffer + 3]
+	rla
+	ld [hMathBuffer + 3], a
+
+	ld a, [hMathBuffer + 2]
+	rla
+	ld [hMathBuffer + 2], a
+
+	ld a, [hMathBuffer + 1]
+	rla
+	ld [hMathBuffer + 1], a
+
+	dec e
+	jr nz, .next2
+
+	ld e, 8
+	ld a, [hMathBuffer + 0]
+	ld [hDivisor], a
+	xor a
+	ld [hMathBuffer + 0], a
+
+	ld a, [hDividend + 1]
+	ld [hDividend + 0], a
+
+	ld a, [hDividend + 2]
+	ld [hDividend + 1], a
+
+	ld a, [hDividend + 3]
+	ld [hDividend + 2], a
+
+.next2
+	ld a, e
+	cp 1
+	jr nz, .okay
+	dec b
+
+.okay
+	ld a, [hDivisor]
+	srl a
+	ld [hDivisor], a
+
+	ld a, [hMathBuffer + 0]
+	rr a
+	ld [hMathBuffer + 0], a
+
+	jr .loop
+
+.done
+	ld a, [hDividend + 1]
+	ld [hDivisor], a
+
+	ld a, [hMathBuffer + 4]
+	ld [hDividend + 3], a
+
+	ld a, [hMathBuffer + 3]
+	ld [hDividend + 2], a
+
+	ld a, [hMathBuffer + 2]
+	ld [hDividend + 1], a
+
+	ld a, [hMathBuffer + 1]
+	ld [hDividend + 0], a
+
+	ret
+; 67c1