tests: run fuzzers four times in a row

This should help to catch issues that are easily detectable by
bad_build_check like the one being fixed inhttps://github.com/systemd/pull/10793,
which would totally break the build tomorrow if I hadn't run
`helper.py check_build` manually.

src/fuzz/fuzz-main.c

@@ -12,6 +12,11 @@
  * It reads files named on the command line and passes them one by one into the
  * fuzzer that it is compiled into. */
 
+/* This one was borrowed from
+ * https://github.com/google/oss-fuzz/blob/646fca1b506b056db3a60d32c4a1a7398f171c94/infra/base-images/base-runner/bad_build_check#L19
+ */
+#define MIN_NUMBER_OF_RUNS 4
+
 int main(int argc, char **argv) {
         int i, r;
         size_t size;
@@ -30,7 +35,8 @@ int main(int argc, char **argv) {
                 }
                 printf("%s... ", name);
                 fflush(stdout);
-                (void) LLVMFuzzerTestOneInput((uint8_t*)buf, size);
+                for (int j = 0; j < MIN_NUMBER_OF_RUNS; j++)
+                        (void) LLVMFuzzerTestOneInput((uint8_t*)buf, size);
                 printf("ok\n");
         }