Relax Dependency Version Requirements #188
Description
Is your feature request related to a problem?
It's generally considered good practice for libraries to have the minimum required version of each project dependency in order to avoid, for lack of a better phrase, dependency resolution hell. If SimOpt is being installed alongside several other libraries, the high/exact version requirements may not play resolve when combined with the dependencies of those other libraries and thus cause the dependencies to not be able to install.
Describe the solution you'd like
I'd like testing to be done in order to confirm the library works with earlier versions of libraries. This might look like testing versions of Numpy earlier than the currently required 2.3.4, figuring out that any version after Numpy 2.0's release will work with the library (and produce the same results), and lowering the minimum required version to 2.0.
Describe alternatives you've considered
If we decide we're ok with a wide dependency version range for deployment but want to use specific versions for development/testing to help isolate issues, we could look into integrating tool-specific lock files (eg uv.lock) into the development process.