vmgs: refactor implementation

[tjones60]

Refactors the vmgs implementation to minimize duplicate code, especially the number of places where the raw VMGS file reads and writes occur.

[Copilot]

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

[chris-oo]

a few comments but mostly reasonable.

What are your thoughts on the whole encryption feature at this point? Should it still exist?

[mebersol]

fn allocate_helper(

might be a good target for unit tests?

---

Refers to: vm/vmgs/vmgs/src/vmgs_impl.rs:1711 in 4b64315. [](commit_id = 4b64315, deletion_comment = False)

[tjones60]

a few comments but mostly reasonable.

What are your thoughts on the whole encryption feature at this point? Should it still exist?

I think the encryption feature is good since it makes the core vmgs implementation more portable. I could envision a future scenario where we are trying to bring up a new platform that doesn't support the crypto we currently have but we want vmgs to work. Plus this allows you to build vmgstool without encryption if that is all you need, which is faster and requires less dependencies.

[tjones60]

fn allocate_helper(

might be a good target for unit tests?

Refers to: vm/vmgs/vmgs/src/vmgs_impl.rs:1711 in 4b64315. [](commit_id = 4b64315, deletion_comment = False)

There are some tests that indirectly test it (test_validate_header, test_initialize_file_metadata) but I can see if I can test it more directly as well.