[ty] Share parameter-default inference between functions and lambdas

[charliermarsh]

Summary

We already infer defaults for named functions separately from their annotations. Call checking only needs to know whether a parameter has a default, and evaluating its value while building the signature can re-enter the decorated function's own signature inference.

We now extend this separation to lambdas, using a shared inference region for each parameter default. Previously, reading a lambda default could infer its entire enclosing scope, causing recursive decorators to grow nested container types without converging.

Closes astral-sh/ty#4613.

[astral-sh-bot]

Typing conformance results

No changes detected ✅

Current numbers

The percentage of diagnostics emitted that were expected errors held steady at 98.24%. The percentage of expected errors that received a diagnostic held steady at 98.24%. The number of fully passing files held steady at 134/146.

[astral-sh-bot]

Memory usage report

Summary

Project	Old	New	Diff	Outcome
prefect	476.02MB	476.40MB	+0.08% (383.15kB)	⏫
sphinx	172.45MB	172.47MB	+0.01% (21.68kB)	⏫
trio	93.09MB	93.10MB	+0.01% (12.09kB)	⏫
flake8	38.29MB	38.30MB	+0.01% (2.29kB)	⏫

Significant changes

Click to expand detailed breakdown

prefect

Name	Old	New	Diff	Outcome
infer_parameter_default_types	0.00B	905.33kB	+905.33kB (new)	⏫
infer_function_default_types	551.86kB	0.00B	-100.00% (551.86kB)	⬇️
infer_scope_types_impl	19.68MB	19.71MB	+0.17% (33.88kB)	⏫
parameter_default_type	26.09kB	24.41kB	-6.44% (1.68kB)	⬇️
infer_definition_types	37.85MB	37.85MB	-0.00% (632.00B)	⬇️
infer_lambda_signature	285.37kB	284.93kB	-0.15% (448.00B)	⬇️
infer_expression_types_impl	24.06MB	24.06MB	-0.00% (432.00B)	⬇️
TypePair	9.02MB	9.02MB	-0.00% (384.00B)	⬇️
UnionType	1.38MB	1.38MB	-0.02% (224.00B)	⬇️
infer_statement_types_impl	674.94kB	674.73kB	-0.03% (216.00B)	⬇️
is_redundant_with_impl	3.10MB	3.10MB	-0.00% (160.00B)	⬇️
union_from_two_elements	222.51kB	222.41kB	-0.04% (96.00B)	⬇️

sphinx

Name	Old	New	Diff	Outcome
infer_parameter_default_types	0.00B	109.55kB	+109.55kB (new)	⏫
infer_function_default_types	91.18kB	0.00B	-100.00% (91.18kB)	⬇️
infer_scope_types_impl	6.20MB	6.21MB	+0.05% (3.32kB)	⏫
parameter_default_type	272.00B	256.00B	-5.88% (16.00B)	⬇️

trio

Name	Old	New	Diff	Outcome
infer_parameter_default_types	0.00B	61.18kB	+61.18kB (new)	⏫
infer_function_default_types	50.33kB	0.00B	-100.00% (50.33kB)	⬇️
infer_scope_types_impl	2.83MB	2.84MB	+0.06% (1.67kB)	⏫
parameter_default_type	5.16kB	4.73kB	-8.32% (440.00B)	⬇️

flake8

Name	Old	New	Diff	Outcome
infer_parameter_default_types	0.00B	7.05kB	+7.05kB (new)	⏫
infer_function_default_types	4.98kB	0.00B	-100.00% (4.98kB)	⬇️
infer_scope_types_impl	536.13kB	536.37kB	+0.04% (240.00B)	⏫
parameter_default_type	72.00B	64.00B	-11.11% (8.00B)	⬇️

[astral-sh-bot]

ecosystem-analyzer results

No diagnostic changes detected ✅

Flaky changes detected. This PR summary excludes flaky changes; see the HTML report for details.

Full report with detailed diff (timing results)

[codspeed]

Merging this PR will not alter performance

✅ 178 untouched benchmarks
⏩ 60 skipped benchmarks¹

---

_{Comparing charlie/fix-ty-4613 (4c829ad) with charlie/ty-stable-lambda-identities (b21c1ae)}

Footnotes

1. 60 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩