Multi-repo project sqlmesh run will only process the first project in the argument list

[sineline]

In a multi-repo project context, when running sqlmesh -p project_1 ... -p project_n run only the first project is processed. However, the following projects get corrupted and will throw 'Error: Duplicate key '"iceberg"."schema"."table_1"' found in UniqueKeyDict. Call dict.update(...) if this is intentional' errors when running individually.

This happens using python 3.12 and sqlmesh 0.226.0

Plans are applied on a single project basis, and the run is handled by a service that attemps to perform the single run against multiple projects. This leads to the behavior described above.