3.0.0: Convert algod responses to typed

[Eric-Warehime]

Resolves #771 for the algod client.

I'm attempting to merge this to the 3.0.0 branch of the SDK which we will not be releasing until we've gained a critical mass of breaking changes that we feel justifies a breaking major version bump for this SDK.

Also updates the eslint dependencies to more recent versions, and fixes a couple of new lints which were failing--optional parameters at the end of the function signature, and promise return in the cucumber tests.

Existing tests pass--not sure if we need to be adding more tests here.

[Eric-Warehime]

@jasonpaulos
Ok I think I've addressed all of the feedback except for the following:

1. Indexer response typing--I've not included them in this PR. It's possible we should include them in this branch before releasing 3.0.0 since it would be a breaking change as well.
2. Converting the storage format of everything to bigint instead of union types. This seems both like a breaking change to me, and also something that's more of an annoyance than anything else. Libraries like https://www.npmjs.com/package/json-bigint seem to solve this problem.

[jasonpaulos]

@Eric-Warehime I realized changing the default int decoding is a bit of a larger problem. I've opened #816 to hopefully address the bigint issue.

If the idea behind that PR makes sense to you, this PR can revert the attempt to change int decoding in favor of that PR.

Everything else here looks good to me.

[Eric-Warehime]

@jasonpaulos

If you want me to I can remove the bigint changes in this PR. I'm not super opinionated on having number | bigint unions w/ the deserialization prefering bigints versus what it looks like you've done in your PR which is to just convert all numbered responses to bigint exclusively.

I guess the caveat there is that you'll need to alter the code generator since you're modifying the types and it looks like your PR has some issues (possibly some of the same test failures I've already worked through here?).

As a final option I guess we could get this merged and follow up with a PR to change the unions to bigint--maybe that would just cause a lot of unnecessary work though.

[jasonpaulos]

I still think our BlockHeader needs to be updated to be correct, but that can be worked out later. Let's get this merged