solana: add more tests

[kcsongor]

No description provided.

[johnsaigle]

More tests are always great! Looks likes this needs a rebase.
After that I'll do a proper review. I'm sure adding tests will only be beneficial though so expect an approval

[kcsongor]

I'll rebase this later today

[johnsaigle]

Looks like this PR has some conflicts to be resolved

[kcsongor]

@johnsaigle rebased!

[johnsaigle]

@kcsongor Clippy is unhappy

[johnsaigle]

EDIT: I committed this change directly to the branch.

---

CI is failing because of existing panic calls in programs/example-native-token-transfers/tests/common/setup.rs.

https://github.com/wormhole-foundation/example-native-token-transfers/blob/cb57ed240bc4e986a144c6bc7e6b7e6fb3823cb5/solana/programs/example-native-token-transfers/tests/common/setup.rs#L420-L425

I think this is somehow related to the rust-toolchain change? If I change to stable the panic is not flagged.

Anyway, I think you can add #[allow(clippy::panic)] just above the failing code block and it should resolve the issue

[johnsaigle]

@kcsongor A couple of notes:

1. I added a commit to the branch: it adds a bad_mint_authority to the TestData object to resolve build failures. I'm not sure if this is the right approach so feel free to nuke that commit if it doesn't make sense to you.

2. I'm very confused about the current test failure on test_receive(). It's getting an Ok(()) from release_inbound and so unwrap_err() panics. However, Ok(()) should never happen when we call try to release an inbox item twice (as it should be in ReleaseStatus::Released and therefore try_release() should throw an error).

Interestingly, the test actually passes if revert_on_delay is enabled (!)
I'm not sure how this is possible. Hopefully you have some insight to figure this one out.

(I opened #336 in the process of trying to debug this. Take a look, maybe it'll help you)