swarm-derive/: Remove support for ignoring fields on struct

[mxinden]

Description

With the removal of NetworkBehaviourEventProcess there is no more need for
ignoring fields.

Any objections? Anyone can still see a use-case for fields that need to be ignored?

Links to any relevant issues

Deprecation of NetworkBehaviourEventProcess #2840 and removal #2840.

Past discussions on the removal of poll #2751 (comment) and #2819 (comment)

Open Questions

Change checklist

• I have performed a self-review of my own code
• I have made corresponding changes to the documentation
• I have added tests that prove my fix is effective or that my feature works
• A changelog entry has been made in the appropriate crates

[thomaseizinger]

No objections!

[dariusc93]

If im reading this right, we can have different fields in the struct without it being specific to any behaviour since originally those fields that didnt implement such behaviour would need to be ignored to be there?

[mxinden]

I am not sure I am following @dariusc93. With this change you can no longer have any fields on your main struct that do not implement NetworkBehaviour. Do you foresee any issues with that?

[dariusc93]

Sorry I wasnt clear (misread the pr when looking over it last night). Ignoring fields do have its benefits. In one of my use case would be to have the struct contain a registry to track QueryId and check against that when receiving a KademliaEvent::OutboundQueryCompleted event and return the results. I know rust-ipfs does the same (though might not be a good example since it havent seen much activities on that side). I might have some workarounds though.

[mxinden]

Ignoring fields do have its benefits. In one of my use case would be to have the struct contain a registry to track QueryId and check against that when receiving a KademliaEvent::OutboundQueryCompleted event and return the results.

Why not store the "registry to track QueryId" in the object (e.g. struct) that is polling the Swarm?