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 83.33333% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 87.33%. Comparing base (4ed06c1) to head (521a683).

Files with missing lines	Patch %	Lines
src/tracksdata/graph/_sql_graph.py	83.33%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #262      +/-   ##
==========================================
- Coverage   87.34%   87.33%   -0.02%     
==========================================
  Files          56       56              
  Lines        4703     4705       +2     
  Branches      829      832       +3     
==========================================
+ Hits         4108     4109       +1     
  Misses        379      379              
- Partials      216      217       +1     

☔ 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.

[JoOkuma]

@yfukai sorry for the delay.
I'll review this once #260 is merged. Everything looks good so far.

[JoOkuma]

Great PR @yfukai,

I appreciate all the tests you're adding.
I left a few minor comments.

[JoOkuma]

@yfukai, I'm a bit puzzled. GitHub diff shows little difference.
Do you think I did something wrong when merging the conflicts?
We can revert the changes.

[yfukai]

Maybe the PR was merged with #259 and now in the repo... I can still fix the code that you commented and continue with this PR! Or instead, we can revert 259 and merge this first.

[JoOkuma]

Gotcha, you can address the comments here.
It's already merged, so we can leave it as it is.
Thanks @yfukai

[yfukai]

Hi @JoOkuma, I'm sorry for the late response! Finally, I could take time on this.
#262 (comment)
About this, I believe this verbose pattern was necessary since we serialize the metadata at the whole metadata level in the SQLGraph. Does this make sense for you?

[JoOkuma]

That makes sense @yfukai, and I'm glad you're back.
I think this PR is ready to merge.
Is that OK for you?

[yfukai]

Hi I'm sorry for the absence! Yes, feel free to merge this!

[yfukai]

I'll continue working on #268