Make magic_setsubstr check UTF8 flag after stringification

By checking it before, it can end up treating a UTF8 string as bytes
when calculating offsets if the UTF8 flag is not turned on until
the target is stringified.  This can happen with overloading and
typeglobs.

This is a regression from 5.14.  5.14 itself was buggy, too, but one
would have to modify the target after creating the substr lvalue but
before assigning to it; and that because of another bug fixed by
83f78d1a27, which was cancelling out this one.

package o {
    use overload '""' => sub { $_[0][0] }
}
my $refee = bless ["\x{100}a"], o::;
my $substr = \substr $refee, -2;
$$substr = "b";
warn $refee;

That prints:

Wide character in warn at - line 7.
Āb at - line 7.

In 5.14 it prints:

b at - line 7.

1 files changed, 3 insertions, 2 deletions

diff --git a/mg.c b/mg.c
index d6561c9c66..7b13f615e3 100644
--- a/mg.c
+++ b/mg.c
@@ -2246,6 +2246,7 @@ Perl_magic_setsubstr(pTHX_ SV *sv, MAGIC *mg)
     dVAR;
     STRLEN len, lsv_len, oldtarglen, newtarglen;
     const char * const tmps = SvPV_const(sv, len);
+    const char *targs;
     SV * const lsv = LvTARG(sv);
     STRLEN lvoff = LvTARGOFF(sv);
     STRLEN lvlen = LvTARGLEN(sv);
@@ -2260,8 +2261,8 @@ Perl_magic_setsubstr(pTHX_ SV *sv, MAGIC *mg)
 	Perl_ck_warner(aTHX_ packWARN(WARN_SUBSTR),
 			    "Attempt to use reference as lvalue in substr"
 	);
-    if (SvUTF8(lsv)) lsv_len = sv_len_utf8_nomg(lsv);
-    else (void)SvPV_nomg(lsv,lsv_len);
+    targs = SvPV_nomg(lsv,lsv_len);
+    if (SvUTF8(lsv)) lsv_len = sv_or_pv_len_utf8(lsv,targs,lsv_len);
     if (!translate_substr_offsets(
 	    lsv_len,
 	    negoff ? -(IV)lvoff : (IV)lvoff, !negoff,