ponder-lab / Hybridize-Functions-Refactoring / #1908

80%

main: 79%

Pull #612

travis-pro

Compute the inferred signature during analysis for eager->hybrid candidates

The eager->hybrid conversion emits the inferred `input_signature` into
the new decorator during the change (`convertToHybrid`), so inference was
not run until then. Compute it during precondition checking too, mirroring
the reconfigure path, so the inferred signature is observable at analysis
time (wizard, evaluator) before the change runs. The result is memoized,
so the change does not recompute it, and computing it does not affect the
P1 decision.

Pin the analysis-time availability in the emission test helper via
`getInferredInputSignature()`.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

6 of 6 new or added lines in 1 file covered. (100.0%)

1543 of 1930 relevant lines covered (79.95%)

0.8 hits per line