Add caching to Secure lookup

[OlafHaalstra]

After having profiled the MFT code it became apparent that the secure.lookup inside get_owner_and_group is the culprit. See image below.

I have compared the following situations:

• Original
• Removing the get_owner_and_group call
• Adding lru_cache to self.lookup
  Which yields the following results:

Situation	Parsing 100.000 records	Parsing 9.206.590 records
Original	1m33s	57m51s
Removed	25s	N/A
Cached	24s	13m27s

Here we can see an improvement of 430% by adding caching.

Steps to create the image:

python -m cProfile -o profile.pstats -m dissect.target.tools.query -t ~/Documents/research/dissect/Collection/ -f mft --limit 100000 | rdump -w test.json
gprof2dot -f pstats profile.pstats | dot -Tpng -o image.out

[OlafHaalstra]

For completeness here is the profile after making the change. Looks like get_owner_and_group still hoards up 41% of the compute time so it's probably very much possible to further improve this.

[Schamper]

By the way, I thought about this and I think if Acquire/Velo would collect the $SII, this issue wouldn't have been as big. Because without it, we have to fallback to a "bruteforce" approach for finding the correct entry.

Created a ticket for Acquire: fox-it/acquire#180

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 87.45%. Comparing base (2765041) to head (321a8ff).

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #35      +/-   ##
==========================================
+ Coverage   87.43%   87.45%   +0.02%     
==========================================
  Files          11       11              
  Lines        1234     1236       +2     
==========================================
+ Hits         1079     1081       +2     
  Misses        155      155              

Flag	Coverage Δ
unittests	87.45% <100.00%> (+0.02%)	⬆️

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

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.