Add lru_cache to many mapper properties

[ml-evs]

This PR is primarily for optimisation of the mapper by lru caching lots of properties on first run. This is still an unfortunate workaround for our weird class-level storage of these properties, but I think this is a reasonable performance fix that maintains backwards-compat.

Need to run some benchmarks before this gets merged.

[codecov]

Codecov Report

Merging #1245 (28e3899) into master (5dc5e4a) will increase coverage by 0.02%.
The diff coverage is 100.00%.

❗ Current head 28e3899 differs from pull request most recent head de1eb51. Consider uploading reports for the commit de1eb51 to get more accurate results

@@            Coverage Diff             @@
##           master    #1245      +/-   ##
==========================================
+ Coverage   90.79%   90.81%   +0.02%     
==========================================
  Files          72       72              
  Lines        4346     4357      +11     
==========================================
+ Hits         3946     3957      +11     
  Misses        400      400              

Flag	Coverage Δ
project	90.81% <100.00%> (+0.02%)	⬆️
validator	90.81% <100.00%> (+0.02%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
optimade/server/mappers/entries.py	97.39% <100.00%> (+0.27%)	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[ml-evs]

Just did some quick tests with the reference server (which has limited data and few provider fields), the validator runs about 10% faster with these changes, so I think we should include this.

[JPBergsma]

I think it is good to add these caches. I suspect there are many more places in the code where this could be useful.