[M] Change in coreboot[main]: commonlib/bsd: Add strnlen() and strlen() functions

[Julius Werner (Code Review)]

Attention is currently required from: Jakub Czapiga, Maximilian Brune, Yu-Ping Wu.

Julius Werner has posted comments on this change by Yu-Ping Wu. ( https://review.coreboot.org/c/coreboot/+/83830?usp=email )

Change subject: commonlib/bsd: Add strnlen() and strlen() functions
......................................................................

Patch Set 1: Code-Review+2

(3 comments)

File payloads/libpayload/libc/string.c:

https://review.coreboot.org/c/coreboot/+/83830/comment/de0c20d8_f6aeeb3d?usp... :
PS1, Line 71: /* NULL and empty strings have length 0. */
            : 	if (!str)
            : 		return 0;
> You don't have this NULL check in the commonlib implementation. […]

There's no NULL check required here by POSIX and other major libc implementations (e.g. glibc: https://github.com/bminor/glibc/blob/master/sysdeps/i386/strlen.c) don't seem to do that. I see no reason why libpayload should do something completely out of turn here, and this seems like as good an opportunity as any to fix that. (It's arguable whether returning 0 is even "better" here. If a caller calls `strlen()` on `NULL` that's always a bug anyway and just silently returning a result that's as incorrect as any other doesn't necessarily make things better.)

File src/commonlib/bsd/string.c:

https://review.coreboot.org/c/coreboot/+/83830/comment/b4e3a6b5_f9174799?usp... :
PS1, Line 12: 	return len;
nit: This is probably overthinking it, but I get a shorter loop in assembly with
```
const char *ptr = 0;
while (*ptr++)
  /* walk string */;
return ptr - str - 1;
```

https://review.coreboot.org/c/coreboot/+/83830/comment/194d8b21_563d6938?usp... :
PS1, Line 20: 	return len;
Here it would be
```
const char *ptr = str;
const char *end = str + n + 1;
while (*ptr++ && ptr < end)
  /* walk string */;
return ptr - str - 1;
```
(I know that makes readability suffer a bit so maybe it's not a good idea. But in general, anything that it can rewrite to check a condition at the end tends to save a jump in the loop. It also seems be able to fit in the post-increment better when it happens at the start (especially on AArch64), and there's no separate counter it could use to calculate offsets instead.)