remove study id from provenance if it has been swapped

[nicoalee]

Steps to reproduce:

1. go to extraction
2. mark a study as save for later or completed
3. swap the study version with another study

Old study is still in the metadata. this means that if the user swaps back to the original study, its status will be saved.

Now that I'm thinking about it...is this a bug or a feature? Do we want the status to be retained when the user swaps back?
If I have study_A_version_B, and then I swap to study_A_version_C, should version_C have the same study status as version_B?