tests: testrunner reliability improvements

- perform torture tests with '-j2' for shorter runtime
- when waiting on test results overly long, log the tests
  waited for and eventually log the test log directories
  for easier analysis what is wrong in CI jobs.
- sockfilt.c: treat the windows errno 109 (ERROR_BROKEN_PIPE)
  as a socket closed by the client and do not exit.
- when verifying https server, do not in addition check
  the http server behind it also
- when tearing down the stunnel of a non-responsive https
  server, tear down the http server with it

Closes #14960

tests/server/sockfilt.c

@@ -218,6 +218,10 @@ static ssize_t write_wincon(int fd, const void *buf, size_t count)
 #define write(a,b,c) write_wincon(a,b,c)
 #endif
 
+/* On Windows, we sometimes get this for a broken pipe, seemingly
+ * when the client just closed stdin? */
+#define CURL_WIN32_EPIPE      109
+
 /*
  * fullread is a wrapper around the read() function. This will repeat the call
  * to read() until it actually has read the complete number of bytes indicated
@@ -243,6 +247,11 @@ static ssize_t fullread(int filedes, void *buffer, size_t nbytes)
       error = errno;
       if((error == EINTR) || (error == EAGAIN))
         continue;
+      if(error == CURL_WIN32_EPIPE) {
+        logmsg("got Windows ERROR_BROKEN_PIPE on fd=%d, treating as close",
+               filedes);
+        return 0;
+      }
       logmsg("reading from file descriptor: %d,", filedes);
       logmsg("unrecoverable read() failure: (%d) %s",
              error, strerror(error));