[SAOC 2025] Separate Semantic Routines from AST Nodes Weekly Update #12

[Mohamed El Shorbagy]

# Summary of Progress (Dec. 01 - Dec. 08)

This week I continued the refactoring process of `mtype` to break 
the semantic dependency on `typesem`. Here are some of the merged 
PRs:

- https://github.com/dlang/dmd/pull/22166
- https://github.com/dlang/dmd/pull/22168
- https://github.com/dlang/dmd/pull/22183
- https://github.com/dlang/dmd/pull/22184
- https://github.com/dlang/dmd/pull/22185
- https://github.com/dlang/dmd/pull/22186
- https://github.com/dlang/dmd/pull/22190
- https://github.com/dlang/dmd/pull/22191
- https://github.com/dlang/dmd/pull/22194

I also found out why some past PRs were failing such as the 
[`deduceWild` PR](https://github.com/dlang/dmd/pull/22107) and 
the [`addSTC` PR](https://github.com/dlang/dmd/pull/22129).

The remaining functions that still depend on `typesem` are 
`nextOf` and `toBaseType`. These functions are used left and 
right across the front-end code, so resolving their dependencies 
will be challenging.