`equzero` and `cmpequ` instrinsics

[ZERICO2005]

Currently, comparing to zero uses __*cmpzero, and checking for only equality between integers uses __*cmpu or __*cmps. Slightly faster routines that only output to the Z flag could be used instead, so I am proposing adding __*equzero and __*cmpeq.

__*cmpzero : Z P
__*equzero : Z

__*cmpu    : Z C
__*cmps    : Z P V
__*cmpeq   : Z

• __*equzero and __*cmpeq can be emitted instead of __*cmpzero and __*cmp(u/s) when only the Z flag (or equality) is needed.
• __*equzero can be aliased to __*cmpzero without issue.
• __*cmpeq can be aliased to __*cmpu or __*cmps without issue.

For the most part, this is mainly a speed optimization. However, adding __llequzero and __i48equzero could be a beneficial size optimization, since the sign flag is basically never used from __llcmpzero and __i48cmpzero. So we could link just the smaller __llequzero and __i48equzero routines.

; 21 bytes
__llcmpzero:
	inc	b
	dec	b
	ret	nz
	inc.s	bc
	dec	c
	jr	nz, .p_nz
	or	a, a
	sbc	hl, bc
	jr	nz, .p_nz
	sbc	hl, de
	ret	z
	add	hl, de
.p_nz:
	inc	b
	ld	b, 0
	ret

; 15 bytes
__llequzero:
	inc	b
	dec	b
	ret	nz
	inc	c
	dec	c
	ret	nz
	add	hl, bc
	or	a, a
	sbc	hl, bc
	ret	nz
	sbc	hl, de
	add	hl, de
	ret

[ZERICO2005]

This can likely be done just as a final peephole optimization checking for if the sign flag was used or not after a call to __*cmpzero.