swarm: Migrate p2p/simulations from go-ethereum

[chadsr]

This PR coupled with this go-ethereum PR migrates p2p/simulations from the go-ethereum repository to here. (Technically it also migrates p2p/testing, but it is already duplicated in this repo)

Motivation for this came about from #1799, where that PR depended on a go-ethereum PR. @acud pointed out that it would in-fact make more sense to migrate p2p/simulations package to ethersphere/swarm.

I refrained from renaming the package, to avoid any confusion. I do feel like it could be refactored to somewhere else though (there are 3-4 simulation related packages scattered around), but I will leave that for core devs to decide. :)

[acud]

@chadsr build is broken

[chadsr]

@chadsr build is broken

@acud Yeah, apologies, my IDE bugged out yesterday and apparently let me down.

However I'm not sure if this PR is a good idea anymore? After finding some complications and what @fjl just added, i'm thinking it might be more appropriate that the packages stay in go-ethereum?

[chadsr]

Going to close this, as it seems best to keep p2p/simulations in go-ethereum after all.

[zelig]

hmm, i thought it was a good idea, given noone is using this and we maintain it...

[chadsr]

Reopening to facilitate some more discussion.
@zelig The package is used by one or two tests in go-ethereum. It seems fjl is keen on keeping the option to use it further in go-ethereum as well.

As an outsider, I see these options (but maybe there are more):

• Continue the migration and use the swarm package as a dependency in go-ethereum.
• Duplicate the package, so that swarm and go-ethereum maintain their own versions. Swarm version could then just be unified with the swarm specific simulation packages that already exist.
• Keep it in go-ethereum and remain reliant on their release cycle.

[acud]

if go-ethereum are using this library then let's not migrate this. let's just PR stuff into go-ethereum and vendor it again. let's just make sure we are not dependent on them to approve this stuff - if we are the codeowners then let's maintain this code (or at least tell geth to vendor this from us so we can actively maintain it)

[acud]

@nonsense @skylenet can I have your input on this? I'd rather we don't maintain duplicate code here and on go-ethereum, and I guess you two would be able to hook it up easily right?

[acud]

@chadsr do we still need this PR? I think due to the fact you've submitted the go-ethereum PR this is no longer needed right?

[chadsr]

@acud Only useful if others from the core team still want to give input, I guess. It's not needed in relation to anything else from me, so feel free to close it.