[WIP] Strict typing and fixed schemas for emmet document models

[esoteric-ephemera]

The goal of this PR is to make fixed, (more-)explicitly typed document models for storing tasks / converting python objects to parquet. To do this, we have to:

• Make document models for some specific pymatgen objects (POTCAR summary stats, ComputedEntry, etc.)
• Make a strict schema (not permitting extra model fields and no union types on fields besides Optional) for TaskDoc
• Maybe concurrent: pydantic models that are low-memory implementations of objects like Structure, PeriodicSite, etc.

To avoid breaking changes, TaskDoc is the same as before, but inherits from a DbTaskDoc which meets the strict criteria above

Other concurrent goals:

• Better PEP585 compliance by partially migrating away from type hints based on the typing package

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 98.55072% with 2 lines in your changes missing coverage. Please review.

Project coverage is 88.99%. Comparing base (e0edac6) to head (0bb8574).

Files with missing lines	Patch %	Lines
emmet-core/emmet/core/tasks.py	97.91%	2 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #1174       +/-   ##
===========================================
+ Coverage   72.13%   88.99%   +16.86%     
===========================================
  Files          77      116       +39     
  Lines        5149    11140     +5991     
===========================================
+ Hits         3714     9914     +6200     
+ Misses       1435     1226      -209     

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