Update Salsa

[MichaReiser]

Summary

This PR updates Salsa to pull in Ibraheems multithreading improvements

salsa-rs/salsa#921

Performance

A small regression for single-threaded benchmarks is expected because papaya is slightly slower than a Mutex<FxHashMap> in the uncontested case (~10%). However, this shouldn't matter as much in practice because:

1. Salsa has a fast-path when only using 1 DB instance which is the common case in production. This fast-path is not impacted by the changes but we measure the slow paths in our benchmarks (because we use multiple db instances)
2. Fixing the 10x slowdown for the congested case (multi threading) outweights the downsides of a 10% perf regression for single threaded use cases, especially considering that ty is heavily multi threaded.

Test Plan

cargo test

[MichaReiser]

@ibraheemdev can you take over this PR. It seems something broke the heap size feature upstream (maybe salsa-rs/salsa#927?)

Edit: I pinned the commit to before said commit for now but that also means that the papaya commit is missing too. We should add a test in upstream salsa that catches the regression and fix it before merging this PR

[github-actions]

mypy_primer results

Changes were detected when running on open source projects

async-utils (https://github.com/mikeshardmind/async-utils)
- TOTAL MEMORY USAGE: ~49MB
+ TOTAL MEMORY USAGE: ~45MB

beartype (https://github.com/beartype/beartype)
- TOTAL MEMORY USAGE: ~88MB
+ TOTAL MEMORY USAGE: ~97MB

discord.py (https://github.com/Rapptz/discord.py)
-     memo fields = ~189MB
+     memo fields = ~207MB

alerta (https://github.com/alerta/alerta)
- TOTAL MEMORY USAGE: ~117MB
+ TOTAL MEMORY USAGE: ~106MB

isort (https://github.com/pycqa/isort)
- TOTAL MEMORY USAGE: ~80MB
+ TOTAL MEMORY USAGE: ~88MB

paasta (https://github.com/yelp/paasta)
-     memo fields = ~156MB
+     memo fields = ~171MB

django-stubs (https://github.com/typeddjango/django-stubs)
-     memo fields = ~156MB
+     memo fields = ~142MB

static-frame (https://github.com/static-frame/static-frame)
-     memo fields = ~304MB
+     memo fields = ~334MB

openlibrary (https://github.com/internetarchive/openlibrary)
-     memo fields = ~171MB
+     memo fields = ~189MB

rotki (https://github.com/rotki/rotki)
-     memo fields = ~445MB
+     memo fields = ~490MB

scikit-learn (https://github.com/scikit-learn/scikit-learn)
- TOTAL MEMORY USAGE: ~717MB
+ TOTAL MEMORY USAGE: ~652MB

[codspeed-hq]

CodSpeed Instrumentation Performance Report

Merging #19020 will degrade performances by 7.34%

_{Comparing micha/salsa-update (baee7c3) with main (ebc70a4)}

Summary

❌ 1 regressions
✅ 38 untouched benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
❌	ty_micro[many_tuple_assignments]	128.9 ms	139.2 ms	-7.34%

[codspeed-hq]

CodSpeed WallTime Performance Report

Merging #19020 will improve performances by ×8.5

_{Comparing micha/salsa-update (baee7c3) with main (ebc70a4)}

Summary

⚡ 1 improvements
✅ 7 untouched benchmarks

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
⚡	multithreaded[pydantic]	2,947.7 ms	346.6 ms	×8.5

[github-actions]

ruff-ecosystem results

Linter (stable)

✅ ecosystem check detected no linter changes.

Linter (preview)

✅ ecosystem check detected no linter changes.

Formatter (stable)

✅ ecosystem check detected no format changes.

Formatter (preview)

✅ ecosystem check detected no format changes.

[MichaReiser]

Hmm, the memory usage changes here must be noise?