Enhancement: Map the cff and codemeta to inveniordm metadata in seperate file

[panta-123]

Description

Looking into the code. My idea is that it will better to seperate the mapping in a seperate file.

cff -> invenioRDM
codemeta -> InvenioRDM

This way we can track the mapping in single place rather than digging through the metadata.py .

And also we can have the check of cff version or codemeta version and its mapping.
Let me know what do you think. I can try to look into it if the idea ia apporved.
Also related to: #38

[tmorrell]

The motivation for the current architecture is that is lets each metadata field get the best value, whether it happens to be in cff or codemeta. If you look at the metadata decisions https://caltechlibrary.github.io/iga/appendix.html#record-metadata, each field has its own rules. I could imaging doing separate mappings, but then you'd have to have another bunch of code to merge the mapped metadata. So I'm not sure which approach would end up being more readable.