Fix GH-15712: overflow on float print with precision ini large value.

When allocating enough room for floats, the allocator used overflows with
large ndigits/EG(precision) value which used an signed integer to
increase the size of thebuffer.
Testing with the zend operator directly is enough to trigger
the issue rather than higher level math interface.

Author: devnexen

Zend/tests/gh15712.phpt

@@ -0,0 +1,9 @@
+--TEST--
+GH-15712: overflow on real number printing
+--FILE--
+<?php
+ini_set('precision', 1100000000);
+echo  -1 * (2 ** -10);
+?>
+--EXPECTF--
+%s

Zend/zend_strtod.c

@@ -3613,11 +3613,11 @@ rv_alloc(i) int i;
 rv_alloc(int i)
 #endif
 {
-	int j, k, *r;
+	int k, *r;
 
-	j = sizeof(ULong);
+	size_t j = sizeof(ULong);
 	for(k = 0;
-		sizeof(Bigint) - sizeof(ULong) - sizeof(int) + (size_t)j <= (size_t)i;
+		sizeof(Bigint) - sizeof(ULong) - sizeof(int) + j <= (size_t)i;
 		j <<= 1)
 			k++;
 	r = (int*)Balloc(k);