RandomX: rewrote dataset read code

Unified code for AMD and Intel
1% faster on Intel
0.15% faster on AMD Ryzen

src/crypto/randomx/asm/program_prologue_win64.inc

@@ -35,6 +35,7 @@
 	mov rbx, r9                 ;# loop counter
 
 	mov rax, rbp
+	ror rbp, 32
 
 	;# zero integer registers
 	xor r8, r8