Randomize test data IDs

[smartspot2]

With how the createtestdata command creates random data, all mentor IDs will match the section IDs; this kind of (very unrealistic) pattern in the test data may cause issues in testing and development, especially when trying to differentiate between mentor and section IDs during development.

While this is being changed, it can also be beneficial to look through other objects to see if any other IDs happen to match up. Test data creation could also be added to, including more edge cases in the data (ex. huge sections).

Making the command perform additional randomization when creating these objects could improve the development process.

[smartspot2]

Resolved by #473