Add implementation for execution token issuance

[longtomjr]

No description provided.

[longtomjr]

There is still some todos that need looking at, and the type names need to be refactored as well. The bulk of the functionality should be here for review though.

[PiDelport]

(Rebased to refresh)

[longtomjr]

TODO: Add in the size of the dataset into the execution token.

[PiDelport]

I rebased to capture the enum naming changes, and tweaked the new imports to be cargo fmt compatible: diff

[PiDelport]

This was pulling in duplications of once_cell-sgx and serde-json-sgx, due to that Cargo issue.

I took a stab at documenting and fixing this a bit better:

• Docs: 4a1fcad
• Cargo.toml changes: a17e368
• Overall diff: 10a3d51..66151a8 (see rtc_auth_enclave/Cargo.lock in particular)

[PiDelport]

(This closes #38, right?)

[longtomjr]

@PiDelport @Nghondzweni . Please have a look at the latest commit diff. I think this wraps up all the feedback, so just double check if there is anything else. Thanks for the reviews!
16084f7

[PiDelport]

@longtomjr: This works, but it might be simpler to use an [out] variable for the buffer capacity used, instead of adding the IssueTokenResponse struct for that?

This seems to be the go-to pattern for variable-sized ECALL out buffers: for example here and here.

I took a stab at amending 16084f7 to use that approach, in 50fb146, and the result looks a bit more like a pattern we'd want to standardise on:

1. EDL: rtc_auth_enclave/rtc_auth.edl
2. Inside enclave: rtc_auth_enclave/src/lib.rs
3. Outside enclave: rtc_uenclave/src/enclaves/rtc_auth.rs

Does that look preferable to you too?

[PiDelport]

Followed up with @longtomjr: happy with the out_token_used parameter. 🚀