abstract NUT-17 subscription kinds

[gudnuf]

closes #279

I made NUT-17 general so that it can easily be extended with new subscription kinds. For now I just defined the proof_state kind in NUT-07 and bolt11_mint_quote/bolt11_melt_quote in NUT-23. We will want the ability to add new kinds for things like new payment methods (ie. bolt12).

For the examples I decided to just leave the concrete examples showing bolt11 of proof states instead of trying to make a general example. The other thing I considered was removing examples from NUT-17 and include the specific examples in the corresponding NUTs, but that makes those NUTs unnecessarily long.

[callebtc]

we can add bolt12 since #259 is now merged

[gudnuf]

we can add bolt12 since #259 is now merged

done

[thesimplekid]

Once open question I had on nut17 and don't think this address is how to handle the info endpoint. Currently the proof state is inside a specific method but a proof isn't tied to a method. Does proof state need to be included in each method?

"nuts": {
    "17": {
      "supported": [
        {
          "method": "bolt11",
          "unit": "sat",
          "commands": [
            "bolt11_mint_quote",
            "bolt11_melt_quote",
            "proof_state"
            ]
        },
      ]
    }
}

[gudnuf]

Does proof state need to be included in each method?

I was thinking that too when doing this. It should be specified outside the methods.

Should we do that in this PR? Will require implementations to change so we should probably keep it in just the bolt11 method for compatibility.

[thesimplekid]

Lets do that separate since this requires no impl changes.

[prusnak]

@thesimplekid can you please resolve the merge conflicts?

@callebtc afterwards this should be OK to merge, ACK?

[thesimplekid]

Rebased.

ACK ca8caf3