RFR(xs): 8225225: stringStream internal buffer should always be zero terminated

[Thomas Stüfe]

Hi all,

please review this small fix and the associated test:

Bug: https://bugs.openjdk.java.net/browse/JDK-8225225
Webrev:
http://cr.openjdk.java.net/~stuefe/webrevs/8225225-stringstream-should-always-be-zero-terminated/webrev.00/webrev/

This patch makes sure the internal buffer of stringStream is always zero
terminated.

It also extends existing gtests to assert this.

In addition, gtests are expanded to test fixed sized stringStream variants
too, and to test for buffer overflow for both bufferedStream and
stringStream (not strictly related to this fix but I was in the area so..).

Thanks, Thomas