Implement a service to deal with Institutions that should be collections

[ManonGros]

Some institutions on GrSciColl should in fact be collections. This seems to be a relatively widespread issue (although I don't have any specific number) and contributes to the issue with duplicate institutions (#189)

Here are 25 examples that I checked manually:
Insitutions_that_should_be_collections_GrSciColl.tsv.zip

• 20 of these institutions have more than one duplicate institution,
• 14 have no collection associated with them,
• 8 have one collection,
• 3 have two collections,
• Collections associated with these institutions don't always match the content of the institution. For example, the collection associated with the USDA/ARS, Southeastern Fruit and Tree Nut Research Laboratory is about Nematodes.
• I could only find obvious replacements for 5 of these institutions, the others would require:
  • either duplicate cleaning beforehand, for example we have many entries for the Cornell University so it would be difficult to pick one for the Cornell University Insect Collection,
  • or the creating of a new institution, for example there is no entry for the "Virginia Institute of Marine Science" for the Virginia Institute of Marine Science, Nunnally Ichthyology Collection.

So one idea could be that it is just a special case of resolving duplicates.
It could be an option of the "replace by" mentioned here: #255
What it would do in addition to everything mentioned in the issue is to create a collection and add it to the institution that will be remaining.
Please share your ideas.

[marcos-lg]

We'll implement a service in the API to convert an institution into a collection.

This service will do:

1. Check that the institution doesn't have collections. Otherwise the service will return an error
2. Create a new collection with the data from the institution. This includes moving the identifiers, machine tags, occurrence mappings and copying the contacts
3. Delete the institution and set a convertedToCollection flag with the key of the new collection
4. The service can receive an institution as a parameter that will be the institution of the new collection. If it doesn't receive this parameter it will create a new institution with the same code as the collection. The name of the new institution must also be specified as a parameter

[marcos-lg]

The endpoint is available now in UAT to receive POST requests at:

http://api.gbif-uat.org/v1/grscicoll/institution/{key}/convertToCollection

In the body it's required to pass either the institution key for the new collection or a name to create a new institution for this new collection. It has to be in JSON format:

{
    "institutionForNewCollectionKey": "dd155a13-33da-46be-9f6e-07809d2ab5ab",
    "nameForNewInstitution": "name"
}

Authentication is required and the user needs to be a GRSciColl admin or editor.

Curl request example:

curl -u username:password -X POST 'api.gbif-uat.org/v1/grscicoll/institution/b779903b-f02b-45f2-91ea-b3b28a0c408e/convertToCollection' \
--header 'Content-Type: application/json' \
--data-raw '{
    "institutionForNewCollectionKey": "dd155a13-33da-46be-9f6e-07809d2ab5ab"
}'

[marcos-lg]

To be changed:

• Don't move the identifiers, just duplicate them and filter the deleted ones in the lookup and in the identifier resolver

[marcos-lg]

I also disallow the conversion iDigBio institutions and restricted the endpoint to GRSCICOLL_ADMINS users only.