Store.connect: fix force_reset kwarg implementations

[rkingsbury]

Summary

This PR increases the consistency of force_reset kwarg in Store.connect() across Store classes. It was motivated by an observation that on systems with slow storage, connecting to a JSONStore can take a long time. If one has already connected to the Store, a subsequent call to connect (with force_reset=False) should not cause the entire connection process to repeat.

This PR enforces the paradigm that we don't need to re-instantiate clients, reset ssh tunnels, or re-read data files if a store has already connected and force_reset=False.

Checklist

• Google format doc strings added.
• Code linted with ruff. (For guidance in fixing rule violates, see rule list)
• Type annotations included. Check with mypy.
• Tests added for new features/fixes.

[codecov]

Codecov Report

Attention: 2 lines in your changes are missing coverage. Please review.

Comparison is base (09f75e2) 87.78% compared to head (2bd123f) 87.76%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #879      +/-   ##
==========================================
- Coverage   87.78%   87.76%   -0.03%     
==========================================
  Files          44       44              
  Lines        3610     3612       +2     
==========================================
+ Hits         3169     3170       +1     
- Misses        441      442       +1     

Files	Coverage Δ
src/maggma/stores/compound_stores.py	81.46% <100.00%> (+0.10%)	⬆️
src/maggma/stores/file_store.py	97.40% <100.00%> (ø)
src/maggma/stores/mongolike.py	87.40% <100.00%> (-0.24%)	⬇️
src/maggma/stores/shared_stores.py	94.11% <ø> (ø)
src/maggma/stores/gridfs.py	94.11% <66.66%> (ø)

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