Automated pipeline

[hurricane642]

Hello everyone,
In this PR, I am adding an automated pipeline for calculations using QE and Perturbo. The idea is to create in advance a folder with input files and a config file to run the calculation correctly. Then all intermediate operations (sequential launching of calculations, creating folders, transferring and creating links to files/folders) are done automatically.
PR is in draft version for now, because I would like to extend the presented functionality. Ideally, I would like to be able to change some calculations parameters (grid, pseudopotentials, etc) directly from pipeline. In addition, I would like you to test its usability.

To run the pipeline, you need a folder similar to the test folder (epr1 for example). Inside, you should have the pw-ph-wann and qe2pert folders, and a config-machine folder with the config_machine.yml file. Its structure looks like this:

COMP_FOLD: results
prefix: gaas
comp_info:
    scf:
        exec: pw.x
    nscf:
        exec: pw.x
    phonon:
        exec: ph.x
    wannier90:
        exec: wannier90.x
    pw2wannier90:
        exec: pw2wannier90.x
    qe2pert:
        prel_coms:
            - export OMP_NUM_THREADS=8
        exec: qe2pert.x

further launch is carried out with the help of entry point command:

auto_pipeline --source_folder epr1

Let me know what do you think about this PR!

[alex2shiyu]

Wonderful feature! Thank you Sergei! I will report back after I enjoy it.