Remove need for representative_periods and graph in add_expression_terms_rep_period_constraints

[abelsiqueira]

Remove graph and representative_periods from add_expression_terms_rep_period_constraints.
Refactor the function to use the DuckDB tables instead of the DataFrames.

This now uses DuckDB to compute the groups and the intersection. The type instability made the code a bit slower so to improve, we have to manually specify the types of some outputs with ::Type. This made the code faster than it was before!

Related issues

Part of #942

Checklist

• I am following the contributing guidelines
• Tests are passing
• Lint workflow is passing
• Docs were updated and workflow is passing

[github-actions]

Benchmark Results

	59d9e96...	7581848...	59d9e96.../7581848a2e217d...
energy_problem/create_model	44 ± 2.9 s	36.1 ± 1.4 s	1.22
energy_problem/input_and_constructor	18.8 ± 0.27 s	18.7 ± 0.11 s	1
time_to_load	3.96 ± 0.014 s	3.97 ± 0.022 s	0.998

	59d9e96...	7581848...	59d9e96.../7581848a2e217d...
energy_problem/create_model	0.563 G allocs: 22.9 GB	0.328 G allocs: 16.5 GB	1.39
energy_problem/input_and_constructor	0.0675 G allocs: 2.37 GB	0.0675 G allocs: 2.37 GB	1
time_to_load	0.159 k allocs: 11.2 kB	0.159 k allocs: 11.2 kB	1

Benchmark Plots

A plot of the benchmark results have been uploaded as an artifact to the workflow run for this PR.
Go to "Actions"->"Benchmark a pull request"->[the most recent run]->"Artifacts" (at the bottom).

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.43%. Comparing base (59d9e96) to head (7581848).
Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #987      +/-   ##
==========================================
+ Coverage   95.37%   95.43%   +0.06%     
==========================================
  Files          29       29              
  Lines        1124     1140      +16     
==========================================
+ Hits         1072     1088      +16     
  Misses         52       52              

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

[datejada]

Thanks @abelsiqueira, I have a couple of comments that are for my understanding of the changes rather than improvements. The SQL is understandable and readable. The best is that it is fast and uses less memory.

[abelsiqueira]

Hi @datejada, thanks for the review and comments. Let me know if my replies answer your questions. I will push updates and re-request a review 👍

[datejada]

@abelsiqueira Thanks for the reply to the comments and the changes. I learnt more about SQL and type stability 😄