Nshm organisational refactor

[lispandfound]

This PR requires ucgmsim/qcore#305 and ucgmsim/qcore#303.

This pull request represents my current vision for how I think the nshmdb should work going forward. The purpose of this tool is as follows:

1. Provide a script to produce an SQLite database to efficiently query fault geometry (and other things going forward).
2. To provide any code relying on NSHM2022 data a simple API to work with.
3. To provide missing visualisation tools for NSHM (essentially, plugging gaps in the rupture explorer).

To this end we now have three modules and one script.

• nshmdb. This module provides an interface to query the database.
• fault. Contains the classes defining a fault geometry.
• plotting.rupture. The plotting package contains extra visualisation tools, for now just a tool to visualise a rupture scenario with pygmt.

The script has also been improved. It now uses the Python zipfile module to read the CRU solution zip file directly (i.e. without the user specifying the paths of the individial files). It does not extract need an extracted zip either, it's all read in binary. This means that the user does not need to understand the CRU solution zip's many undocumented files to generate the database. It is as simple as downloading the CRU solution zip file and running:

$ python nshmdb/scripts/nshm_db_generator.py ~/Downloads/CRU_fault_system_solution.zip nshmdb.db

or if they've installed the package the script lives in their path so this also works

$ nshm-db-generator ~/Downloads/CRU_fault_system_solution.zip nshmdb.db

The script now gives live updates via tqdm on the database build progress (which varies from 5 minutes on my home desktop, to 15 minutes on my work desktop).

Here is an example of how you can use the new API:

>>> from nshmdb import nshmdb
>>> from nshmdb.plotting import rupture
>>> db = nshmdb.NSHMDB('nshmdb.db')
>>> rupture_100 = db.get_rupture_faults(100)
>>> akatarawa = rupture_100[0]
>>> plane = akatarawa.planes[0]
>>> np.round(plane.dip)
75.0
>>> plane.bottom_m
20000.0
>>> rupture.plot_rupture('Rupture 100', rupture_100)
# see plot below

The above code produces the following plot of the faults involved in the rupture.

[joelridden]

Decent work! Not that strong on some of the math side for some of the fault calculations but I'm sure it works. Just a few minor comments.

Love the documentation

[lispandfound]

we can probably merge this after @claudio525 and @sungeunbae approve.

[lispandfound]

Sorry @claudio525, just had to switch to optional so it works on nesi 🫤