draft: metadata support for columns

[vincentkoppen]

For #136 we would like to have a programmable way to recognize id columns.

I tried by setting the type, however I did not really get anywhere with this. Instead I went with the metadata route in this setup.
It works, but not to sure whether I/we are comfortable with this implementation. Specifically since numpy has the following warning ;)

https://numpy.org/doc/stable/reference/generated/numpy.dtype.metadata.html

Although used in certain projects, this feature was long undocumented and is not well supported. Some aspects of metadata propagation are expected to change in the future.

How it works:

• We add support to annotate a NDArray within a dtype definition with a dict (or we can choose a different datastructure)
• We set the metadata provided in the dtype
• We can now access the metadata where we want.

The main.py example code has the following output:

[('id', '<i4'), ('measured_object', '<i4'), ('u_sigma', '<f8', (3,)), ('u_measured', '<f8', (3,)), ('u_angle_measured', '<f8', (3,))] #the dtype
{'id': True} #metadata of "id" in dtype

[sonarqubecloud]

Quality Gate failed

Failed conditions
79.1% Coverage on New Code (required ≥ 80%)

See analysis details on SonarQube Cloud