Back-end sends new information for cluster load indicator

[MartinBelthle]

Problem Statement

Currently the cluster load indicator is as such :

This indicator is retrieved by the front-end via the end-point GET v1/launcher/load which returns a dict :
{'slurm': float, 'local': float}

With this PR : #1664, the back-end will send more accurate information with another format : LauncherLoadDTO().
This class has 4 variables which are not optional:

• allocated_cpu_rate: float,
• cluster_load_rate: float,
• nb_queued_jobs: int,
• launcher_status: str

The desired outcome

The front-end will have to do 2 things:

• Display the first 3 indicators (all except launcher_status) at the same place it used to be. You are free to choose the mockup you want.
• If launcher_status != "SUCCESS": it means that the cluster load could not be retrieved. The first 3 indicators will be at 0. Probably just putting 0 is not enough to show that the info could not be retrieved but I do not have a precise idea on how to display it in the front.

Developer Notes

• You can do your development on the PR linked above because it returns the new object LauncherLoadDTO().
  NB : Probably use the branch on this commit : d549e16 because after this the UTs do not pass anymore