Configurable inlining heuristics

It would be useful to support a configurable level of heuristics, such that a user could tune down inlining (prioritizing bundle size). Architecturally, this means that we would need to support customization of our various heuristic guards, which are currently hard coded.

Start by locating the hard-coded heuristic guards that control inlining; no files, tests, or entry points are named in the issue. Define how users would configure the heuristic level and verify that tuning it down prioritizes smaller bundles while preserving correct optimization behavior.

Written by the indexing model from the issue text.