Saving dtypes as metadata for roundtrip consistency

[yfukai]

This is a follow-up on the dtype, assuming #260 has been merged.
In this PR, the graphs store the dtypes of the fields as metadata, enabling robust recovery of the original dtypes during round-trip from_other conversions.

[yfukai]

Notes:

• I should also store the default values. Serializing the polars dataframes is useful to store the values?
• The roles of AttrSchema overlap with the current metadata. This could be SQLGraph-specific machinery.

[yfukai]

Both tasks are done. I'll review the code for SQLgraph and mark this open for review.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 81.74274% with 44 lines in your changes missing coverage. Please review.
✅ Project coverage is 87.70%. Comparing base (e10d8c3) to head (cc55976).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/tracksdata/graph/_base_graph.py	72.97%	18 Missing and 2 partials ⚠️
src/tracksdata/utils/_dtypes.py	71.83%	11 Missing and 9 partials ⚠️
src/tracksdata/graph/_graph_view.py	66.66%	2 Missing ⚠️
src/tracksdata/graph/_sql_graph.py	98.57%	1 Missing ⚠️
src/tracksdata/nodes/_regionprops.py	50.00%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #262      +/-   ##
==========================================
- Coverage   88.18%   87.70%   -0.48%     
==========================================
  Files          56       56              
  Lines        4308     4531     +223     
  Branches      749      781      +32     
==========================================
+ Hits         3799     3974     +175     
- Misses        319      356      +37     
- Partials      190      201      +11     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.