Remove CCE dependency from Interpolation

[knelli2]

Proposed changes

Addresses the CCE part of #3561.

Moves SendGhWorldtubeData to CCE.

Upgrade instructions

src/ParallelAlgorithms/Interpolation/Callbacks/SendGhWorldtubeData.hpp has been moved to src/Evolution/Systems/Cce/SendGhWorldtubeData.hpp

Code review checklist

• The code is documented and the documentation renders correctly. Run
  make doc to generate the documentation locally into BUILD_DIR/docs/html.
  Then open index.html.
• The code follows the stylistic and code quality guidelines listed in the
  code review guide.
• The PR lists upgrade instructions and is labeled bugfix or
  new feature if appropriate.

Further comments

[nikwit]

maybe it would be more logical to move the callback into CCE itself? By forward declaring it, it seems like it can be used without CCE but if the executable does not have CCE as a link target, you will end up with an undefined symbol, or am I not seeing it right?

[knelli2]

You're right, but that's true of any forward declaration and we use forward declarations in a lot of places. I guess it's just a preference on whether we want to keep all the interpolation callbacks in the same place or not. I don't see an obvious reason to require them to be in the same place.

Furthermore, up until now, we haven't had any callbacks that truly depended on other libs. But in theory you could write a callback that does depend upon other libs. So in that case, you'd want to store it elsewhere. I think that's enough reason for me to move it to CCE. Does that sound good @nikwit?

[nikwit]

I think the reason we use forward declaration is mostly to reduce compilation time so we only include the symbols that we need rather than the entire header. Correct me if I'm wrong but the forward declaration here does not do much in terms of dependencies and I suspect just deleting the link dependency from CMake would also compile and link fine. I agree we should put it into CCE to actually get rid of the dependency and so that the symbol can not exist without its definition.

[knelli2]

@nikwit ok it has been moved!

[nikwit]

Looks good, but I am not sure if we shouldn't also move it into the CCE namespace? Whaat do you think?

[knelli2]

We could, but currently this isn't actually used anywhere except a test so not super important. I think I'll leave it as is for now