fallback to Scheduler if the GRID is not available on runtime

[rvhonorato]

Checklist

• Tests added for the new code
• Documentation added for the code changes
• Modifications / enhancements are reflected on the haddock3 user-manual -> add explanation about GRID fallback haddock3-user-manual#26
• CHANGELOG.md is updated to incorporate new changes
• Does not break licensing
• Does not add any dependencies, if it does please add a thorough explanation

Summary of the Pull Request

This PR adds a fallback to get_engine to check if the GRID connection is possible before returning GRIDScheduler.

The fallback logic is to call libgrid.ping_dirac that will check if the necessary commands are available in the PATH and also run dirac-proxy-info that will return a non-zero code if the connection has not been made.

The behavior is then: if an user sets mode=grid but the grid is not available, it will use the local processors instead of the grid, inhering whatever ncores was defined in the configuration file and also display a warning in the logs.

The check will be done per-module, so it should capture the scenario in which one module can finish with the grid but for some reason it becomes unavailable, the following will then use the multiprocessing scheduler. This should give is a higher execution fault tolerance.

Related Issue

#1472

[VGPReys]

Seems good