gnrc_netif: allow for wait of minimum time between sends

[miri64]

Contribution description

This was cherry-picked out of #11068 (which became a huge mess during my experiment phase in preparation for the 6LoWPAN fragment forwarding paper.

This adds a minimum sleep between every send, to allow for slowing down the send operations of a device.

To be perfectly honest, I don't know if and how useful this could be in production operation or beyond experimentation in general, but it helped me to confirm some things during my experimentation. However, I kept it optional and marked it as experimental so I think it does not harm to merge.

Testing procedure

First compile and flash gnrc_networking normally with GNRC_NETIF_MIN_WAIT_AFTER_SEND_US=0. Ping another node with the same configuration. There should be no significant changes to to master.

Now compile with a higher delay e.g. GNRC_NETIF_MIN_WAIT_AFTER_SEND_US=5000. The RTT when pinging should now significantly increase (maybe even timeout), if you have access to a sniffer: the time different between two transmissions (ignoring L2 retransmissions) should now be at minimum at the configured time interval.

Issues/PRs references

Cherry-picked from #11068, but this PR has no dependencies.

[fjmolinas]

Tested using a samr21-xpro and pba-d-01-kw2x. I only have a minor comment on the doc, if you agree please squash directly.

• GNRC_NETIF_MIN_WAIT_AFTER_SEND_US=5000

> ping6 fe80::d1c1:6d4e:ab6a:1336 
2019-08-09 10:07:59,059 - INFO #  ping6 fe80::d1c1:6d4e:ab6a:1336
2019-08-09 10:08:02,058 - INFO # 
2019-08-09 10:08:02,062 - INFO # --- fe80::d1c1:6d4e:ab6a:1336 PING statistics ---
2019-08-09 10:08:02,067 - INFO # 3 packets transmitted, 0 packets received, 100% packet loss

• GNRC_NETIF_MIN_WAIT_AFTER_SEND_US=500

ping6 fe80::d1c1:6d4e:ab6a:1336 
2019-08-09 10:08:58,750 - INFO # ping6 fe80::d1c1:6d4e:ab6a:1336
2019-08-09 10:08:58,766 - INFO # 12 bytes from fe80::d1c1:6d4e:ab6a:1336: icmp_seq=0 ttl=64 rssi=-57 dBm time=8.689 ms
2019-08-09 10:08:59,765 - INFO # 12 bytes from fe80::d1c1:6d4e:ab6a:1336: icmp_seq=1 ttl=64 rssi=-63 dBm time=8.995 ms
2019-08-09 10:09:00,764 - INFO # 12 bytes from fe80::d1c1:6d4e:ab6a:1336: icmp_seq=2 ttl=64 rssi=-64 dBm time=8.371 ms
2019-08-09 10:09:00,764 - INFO # 
2019-08-09 10:09:00,768 - INFO # --- fe80::d1c1:6d4e:ab6a:1336 PING statistics ---
2019-08-09 10:09:00,774 - INFO # 3 packets transmitted, 3 packets received, 0% packet loss
2019-08-09 10:09:00,778 - INFO # round-trip min/avg/max = 8.371/8.685/8.995 ms

• GNRC_NETIF_MIN_WAIT_AFTER_SEND_US=0

2019-08-09 10:06:41,414 - INFO #  ping6 fe80::d1c1:6d4e:ab6a:1336
2019-08-09 10:06:41,429 - INFO # 12 bytes from fe80::d1c1:6d4e:ab6a:1336: id=83 seq=1 hop limit=64 time = 7.684 ms
2019-08-09 10:06:42,443 - INFO # 12 bytes from fe80::d1c1:6d4e:ab6a:1336: id=83 seq=2 hop limit=64 time = 7.692 ms
2019-08-09 10:06:43,457 - INFO # 12 bytes from fe80::d1c1:6d4e:ab6a:1336: id=83 seq=3 hop limit=64 time = 8.321 ms

[fjmolinas]

ACK, please squash.

[miri64]

ACK, please squash.

Done

[fjmolinas]

All green! GO!