Confusing time reported by hsbench

It would be good if hsbench also reports the time spent for a single iteration.

As per the default (of doing 20 iterations) with the following results:
```
Time spent scanning: 5.270 seconds
Corpus size: 177,087,567 bytes (130,957 blocks)
Matches per iteration: 637,380 (3.686 matches/kilobyte)
Overall block rate: 496,998.94 blocks/sec
Mean throughput (overall): 5,376.56 Mbit/sec
Max throughput (per core): 5,386.73 Mbit/sec
```
if you quickly do the match yourself (scanning 177 MB in 5.2 seconds, you would get about 34 MB/sec or 272 Mbit/sec), so a wildly different result ....

By passing in the `-n 1` parameter you get the time for a single loop:
```
Time spent scanning: 0.270 seconds
Corpus size: 177,087,567 bytes (130,957 blocks)
Matches per iteration: 637,380 (3.686 matches/kilobyte)
Overall block rate: 485,596.94 blocks/sec
Mean throughput (overall): 5,253.22 Mbit/sec
Max throughput (per core): 5,253.26 Mbit/sec
```
and then it all makes sense.

So it would be nice to also report the time for a single iteration, so as to avoid confusion.