Automatic double grid convergence using IP calculations in frequency space

[palful]

I added two new classes, yambopy/double_grid/dg_setup.py and yambopy/double_grid/dg_convergence.py, as well as a new tutorial, for the convergence of the double grids.

• dg_setup.py is used to generate an rt-compatible "FixSymm" yambo SAVE.
• dg_convergence.py is a workflow that runs multiple qe-nscf and yambo-ip calculations following a tree structure where for every coarse grid provided, a series of "children" fine grids are automatically computed. New coarse and fine grids can be added along the way. For each case, IP absorption is computed.

What is missing: automatic analysis of the momenta of the spectral functions to determine convergence of fine grids.

[palful]

The workflow has been consistently updated in order to support:

• Double grid workflow for any type of yambo calculations (not just ip)
• Cluster submission with parallel job management

[palful]

Another change to be implemented:

• Instead of creating multiple SAVE folders for each fine grid, use only one. Place the the double grid database for each fine grid in a dedicated directory other than SAVE, to be called with -J when executing yambo.