bdiff: don't check border condition in loop

`plast = a + len - 1`. So, this "for" loop iterates from "a" to "plast",
inclusive. So, `p == plast` can only be true on the final iteration
of the loop. So checking for it on every loop iteration is wasteful.

This patch simply decreases the upper bound of the loop by 1 and
adds an explicit check after iteration for the `p == plast` case.
We can't simply add 1 to the initial value for "i" because that
doesn't do the correct thing on empty input strings.

`perfbdiff -m 3041e4d59df2` on the Firefox repo becomes significantly
faster:

! wall 0.072763 comb 0.070000 user 0.070000 sys 0.000000 (best of 100)
! wall 0.053221 comb 0.060000 user 0.060000 sys 0.000000 (best of 100)

For the curious, this code has its origins in 8b067bde6679, which is
the changeset that introduced bdiff.c in 2005.

Also, GNU diffutils is able to perform a similar line-based diff in
under 20ms. So there's likely more perf wins to be found in this code.
One of them is the hashing algorithm. But it looks like mpm spent
some time testing hash collisions in d0c48891dd4a. I'd like to do the
same before switching away from lyhash, just to be on the safe side.

mercurial/bdiff.c

@@ -31,8 +31,10 b' int bdiff_splitlines(const char *a, ssiz'
 
 	/* count the lines */
 	i = 1; /* extra line for sentinel */
-	for (p = a; p < a + len; p++)
-		if (*p == '\n' || p == plast)
+	for (p = a; p < plast; p++)
+		if (*p == '\n')
+			i++;
+	if (p == plast)
 			i++;
 
 	*lr = l = (struct bdiff_line *)malloc(sizeof(struct bdiff_line) * i);