Add ability to create custom model and pass them in to extract

[saulshanabrook]

This PR closes #241 by adding support for custom models in Python.

[codspeed-hq]

CodSpeed Instrumentation Performance Report

Merging #357 will degrade performances by 17.08%

_{Comparing custom-cost-model (9aa04b1) with main (a46b180)}

Summary

❌ 1 (👁 1) regression
✅ 6 untouched

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
👁	test_jit[lda]	7.1 s	8.6 s	-17.08%

[saulshanabrook]

@codex can you review this

[Copilot]

Pull Request Overview

This PR adds support for user-defined cost models in Python by implementing custom extractors that can use Python functions to determine extraction costs. The change enables users to define cost models as Python callable protocols that take an egraph, expression, and children costs, returning the total cost for extraction.

Key changes:

• Implement custom cost model infrastructure with CostModel protocol and Extractor class
• Add Value type to bindings for representing egglog values
• Simplify TermDag implementation using tuple struct pattern

Reviewed Changes

Copilot reviewed 26 out of 27 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
src/extract.rs	New module implementing custom cost models and extractors with Python bindings
src/egraph.rs	Add Value type and value conversion methods for different egglog sorts
src/termdag.rs	Refactor to use tuple struct pattern and remove redundant From implementations
python/egglog/egraph.py	Add cost model protocols, extraction overloads, and default/greedy cost models
python/egglog/declarations.py	Add GetCostDecl and ValueDecl expression types for cost model support

Comments suppressed due to low confidence (1)

src/egraph.rs:1

• Debug print statement should be removed or converted to proper logging.

// Wrapper around EGraph type

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

[saulshanabrook]

@actions-user changelog

[saulshanabrook]

Thanks @FTRobbin and @oflatt for the idea of how to encode graph extraction in the cost model!