KerberosString should use UTF-8 by default

In earlier version of Java kerberos, KerberosString was treated as pure ASCII and stored and loaded as a GeneralString. In JDK-6862679, we realized Windows is using UTF-8. A system property (sun.security.krb5.msinterop.kstring) was added so that when it's set to true UTF-8 is used. We keep the default encoding to ASCII to be compatible.

In fact, RFC 3961 requires it to be UTF-8 string, and MIT krb5 also uses UTF-8. It's time to change the default to UTF-8. The system property can be kept.