Settle the question of SST definition in JEDI obs naming conventions

## Description

There is an open question about naming in the sea surface temperature processing. When we did the naming convention for JEDI we defined `seaSurfaceTemperature` as a variable. This is fine for our current processing as in situ temperature and potential temperature are identical at the surface. However, with the change to conservative temperature, the temperature is not the same as in situ temperature at the surface. Therefore we perhaps should add a `seaSurfaceConservativeTemperature` variable name.

In the SST configurations this is dealt with by overwriting the in situ temperatures with conservative temperature. This works adequately but is not particularly transparent. An alternative would be to not use `seaSurfaceTemperature` at all and just use `waterTemperature` and `waterConservativeTemperature` variable naming in the SST configurations, but this requires significant changes in the yaml.

This issue should be cross-referenced with a similar change in orca-jedi and jjdocs.

The issue was identified in http://fcm9/projects/utils/ticket/629

## Acceptance Criteria (Definition of Done)

Settle a consistent approach to conservative sea surface temperature.

Start by reviewing the SST configurations and the current use of seaSurfaceTemperature, waterTemperature, and waterConservativeTemperature. Compare the alternatives described in the issue and cross-reference the corresponding changes in orca-jedi and jjdocs. Done means a consistent approach to conservative sea surface temperature is agreed and reflected across the relevant projects.

Written by the indexing model from the issue text.