fixes New api

[rienafairefr]

I already did some stuff but didn't push it. :-|

summary:

• to get a deployment result, in the old API it was in the experiment JSON, in the new API it's a separate endpoint, I added a iotlab-experiment deployment subcommand
• in the old API, iotlab-profile addm3 with just the name works, in the new API, it fails with 500
• in the old API, iotlab-profile addm3 when the profile already exists works without errors, in the new API it fails with code 500

integration tests are not yet fully passing, but it should be OK by the end of the day, I'll update this PR

[rienafairefr]

update:

integration tests are passing. In the new API some endpoints (I saw it on DELETE /profiles/{name}) can return non-200 codes that are not errors, for exemple deleting a profile succesfully returns 204. cli-tools framework can only handle 200 and 500 as we've already discussed. I added a workaround for that particular endpoint. we'll have to do the same for other endpoings (like mobility, I think)

I'll add something to recover the coverage and then it's OK for me

[rienafairefr]

update:

• all tests OK, integration tests also. when the API returns 204, then the CLI shows that, no content:

  (cli-tools) matthieu@zurkus:~/iot-lab-dev/parts/cli-tools (new_api)$ iotlab-profile addm3 -n test_m3 -sniffer -channels 11
  {
  "nodearch": "m3", 
  "power": "dc", 
  "profilename": "test_m3", 
  "radio": {
      "channels": [
          11
      ], 
      "mode": "sniffer", 
      "num_per_channel": null, 
      "period": null
  }
  }
  (cli-tools) matthieu@zurkus:~/iot-lab-dev/parts/cli-tools (new_api)$ iotlab-profile del -n test_m3
  (cli-tools) matthieu@zurkus:~/iot-lab-dev/parts/cli-tools (new_api)$

[rienafairefr]

Tested the robot endpoints visible through iotlabcli, seems ok.

@fsaintma there might be a edge case bug in the API with update mobility: when sending node list [] it's doing the update on all the robots in the experiments.

[rienafairefr]

:+1: seems good to merge for future v 3.0 release

[fsaintma]

@rienafairefr ok for None in the case of 204 return. Should you resolve the rebase conflicts ? It seems that you have some commits of me in this PR :) After I will merge it.

[rienafairefr]

Yep, this branch had diverged in a different place. I've rebased on top of your upstream/new_api branch and all seems good now

[rienafairefr]

I recommend a squash, there's a lot of small fix commits ^^