Add minimal documentation on using SHACL and ShEx for Validation

[mfhepp]

I just realized that there are now regularly updated SHACL and ShEx shape definitions for validating data included in each release, like

• https://github.com/schemaorg/schemaorg/blob/main/data/releases/26.0/schemaorg-shapes.shacl
• https://github.com/schemaorg/schemaorg/blob/main/data/releases/26.0/schemaorg-subclasses.shacl

plus

• the schema.org vocabulary with http-based URIs as schemaorg-current-http.ttl and with https-based URIs as schemaorg-current-https.ttl and
• http-to-https mappings in the form of httpequivs.ttl.

It would be great if anybody could provide a super-short how-to on how these are meant to be used, e.g. with pySHACL.

1. Which degree of inferencing is needed (RDFS, OWL2-RL)?
2. Assuming we have two RDFlib graph objects, data and shapes, what is the intended workflow?

I envision something like this:

import rdflib
from pyshacl import validate

# Which parts should go into the shape graph?
shape_components = [] # List of URLS
rules = rdflib.Graph()
for uri in shape_components:
    rules.parse(url, format="turtle")
# Which parts beside my data should go into the data graph?
data_components = [] # List of URLS or files
data  = rdflib.Graph()
for uri in data_components:
    data.parse(url, format="turtle")
# Which level of inferencing is expected? 
# none, rdfs, or owlrl?
status, results, text = validate(data, shacl_graph=rules, inference='none', serialize_report_graph=False)

But since the shape definitions reference schema-specific types, this leads to lots of violations. The http/https issue adds complexity, too.

So any help will be very much appreciated ;-)!

I am pretty sure many others would benefit.

/CC @danbri @ericprud @ashleysommer