RPCCost (RHPv2)

[peterjan]

This PR adds the rpcCost type from hostd. We can move it over to core and get rid of some of the very rough estimates for proof sizes and replace them by exact values we get from DiffProofSize.

[lukechampine]

I'm not entirely sure why the RHP stuff in v2 is different. In any case, the whole v2 package is gone in its-happening -- I decided to take a different approach.

I think my preference is:

• Implement exact cost calculations in core now (for use in RHPv2/RHPv3)
• Use fixed costs in RHPv4

[peterjan]

Yeah I saw you removed v2 in your branch, that is why I thought you no longer wanted to use DiffProofSize and pass in the exact RPCWriteActions to the functions calculating the cost of the RPC. I pushed a small commit that reverts my changes to the v2 package and changes the function signatures a bit.

However, since you mention we ideally go for exact cost calculations. Would you prefer changing some of those functions entirely to accept []RPCWriteAction and implement DiffProofSize as you did in the v2 package? I might go ahead and try to implement that already to see the difference.

[peterjan]

For rhpv3 we could/should update the cost methods on the HostPriceTable to take an argument that indicates whether a proof is required and then also all necessary information to calculate the proof size. Is that something we want to add as well?

[peterjan]

@n8maninger hostd stopped doing it so I copied that behaviour, just wanted to draw attention to it to ensure it's safe to do so

[peterjan]

Functionally this type matches the ResourceCost in RHPv3, we could move this into the types package maybe and use the one type. I didn't want to do that yet mostly because of the documentation in ResourceCost which is tailored to the MDM... Maybe it's fine to just have both, for the same reason hostd had its own internal rpcCost type.

[ChrisSchinnerl]

*LeafSize