Use network-agnostic Address for encoded contract scripts

[kettlebell]

Close #88

[coveralls]

Pull Request Test Coverage Report for Build 2865742887

• 75 of 176 (42.61%) changed or added relevant lines in 14 files are covered.
• 7 unchanged lines in 4 files lost coverage.
• Overall coverage increased (+0.5%) to 65.692%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
core/src/oracle_config.rs	0	1	0.0%
core/src/pool_commands.rs	0	1	0.0%
core/src/main.rs	0	2	0.0%
core/src/cli_commands/extract_reward_tokens.rs	0	3	0.0%
core/src/cli_commands/update_pool.rs	1	4	25.0%
core/src/cli_commands/transfer_oracle_token.rs	0	5	0.0%
core/src/cli_commands/vote_update_pool.rs	1	7	14.29%
core/src/cli_commands/bootstrap.rs	1	9	11.11%
core/src/oracle_state.rs	0	8	0.0%
core/src/cli_commands/prepare_update.rs	3	12	25.0%

Files with Coverage Reduction	New Missed Lines	%
core/src/box_kind/ballot_box.rs	1	88.39%
core/src/cli_commands/bootstrap.rs	1	86.53%
core/src/cli_commands/update_pool.rs	1	71.3%
core/src/main.rs	4	0%

Totals
Change from base Build 2838643233:	0.5%
Covered Lines:	1955
Relevant Lines:	2976

---

💛 - Coveralls

[kettlebell]

@greenhat this is ready for review if you want to merge this before PR #80.

[greenhat]

I glanced through the code and found that OracleConfig::on_mainnet can be safely removed as well. Also, there are still way too many places using AddressEncoder. Could you please take care of these?

[greenhat]

Also BootstrapConfig::on_mainnet can be safely removed as well.

[kettlebell]

Sure will do.

[kettlebell]

It was crucial that OracleConfig::oracle_address is encoded as it allows us to obtain the network prefix used to generate ballot_token_owner_address_str, since we no longer have OracleConfig::on_mainnet.

[kettlebell]

It's ready for review. One addition I made is to check for consistency of network prefixes in more places, so it will error-out if the operator tries to mix testnet and mainnet addresses.

[greenhat]

Looking great! Thank you!
My bad for insisting on Address vs. NetworkAddress. I believe we can ditch more network prefixes. See my comments.

[greenhat]

I think that eventually, we should switch to base16 serialized ErgoTree for storing contracts in the config files. I mean, the odds of someone setting the new contracts with a testnet prefix are super tiny but non-zero.
Of course, it's out of the scope of this PR.

[kettlebell]

Made an issue for this: #98

[greenhat]

I also found a hardcoded mainnet in main::get_change_address_from_node(). Could you please take care of this (switch to unchecked parsing)?

[kettlebell]

I've addressed all your comments.

[greenhat]

Great job! Thank you!