Improve efficiency in source code scanning

[asafgabai]

Is your feature request related to a problem? Please describe.
The applicability scanner operates individually for each module (except for Maven), scanning all files within its root path. In cases where modules are nested within one another, the scanner may inadvertently scan certain files multiple times. In contrast, the other source code scanners run once for the entire project, using IntelliJ's base path as the project root, even if the modules are located outside of this directory.

Describe the solution you'd like to see
The scanners should get a list of roots (based on the modules' paths) that don't contain each other. For all scanners except the applicability scanner, the existing practice of obtaining IntelliJ's base path should be retained alongside the new module-based root list.