feat: add block proposer to addresses_with_balance_change

[joel-u410]

Per this comment on #199, block proposers also need balances fetched.

Figured this was an easy change -- @Fraccaman here's a PR implementing it, assuming this is the right fix here.

[joel-u410]

Hmm, actually not sure this is right -- I'm seeing proposer_address like cafad8da813bae48779a4219a74632d5dca49737 (which I think is the consensus key / tendermint address) but it probably needs to be e.g. tnam1qyctcwkgthr06k7lx38zmjka5dakmvhhyyr0zafu (the namada validator address).

Is there an easy way to get to the tnam address from the tendermint key/addr? It looks like the become_validator transactions have both an address and consensus_key field, but that doesn't cover the genesis validators, and the key is in a different format (but I'm sure could be converted).

Update: I see namadac can do it via the find-validator command. I'll take a look at what it's doing and see if it can be replicated in namada-indexer. Ultimately, it would be great to store both of these addresses in the validators table for every validator.

[Fraccaman]

you are right! we can only query balances given a namada address. I think you can fetch the proposal then call

RPC.vp()
                    .pos()
                    .validator_by_tm_addr(context.client(), &tm_addr)
                    .await` to get the corresponding namada address

Agree it would be better to store these in a separate table but not a big deal for now.

[joel-u410]

Should this PR target 1.0.0-maint?

[Fraccaman]

yep! thanks

[joel-u410]

This compiles, but doesn't quite seem to be working yet (I'm getting None for the validator address) -- will mark ready when I've confirmed it's working.

[joel-u410]

Ah, I think I've solved it -- turns out the RPC is case sensitive, and tm_addr must be uppercase (but is lowercase in shared::block::Block!)

[joel-u410]

After rebasing to 1.0.0-maint, this is going to have merge conflicts with main -- though it should be relatively straightforward to resolve. I kept the pattern of combining HashSets into all_balance_changed_addresses here, in anticipation of merging to main where it also has pgf_receipient_addresses. One thing to watch out for is that here I used .iter().chain() instead of .union() which will make it simpler to combine the 3 HashSets when this is merged to main. So, when merging, you should keep this .iter().chain() version and simply add the line

        .chain(pgf_receipient_addresses.iter())