follow-up(miden-proving-service): address missing comments

[SantiagoPittella]

This PR aims to address the comments of this review.

I tried to maintain commits as we do in reviews, so the reviewer can keep track of which comment I'm addressing in each one.

It:

• Removes the testing feature from the proving-service crate.
• Updates README's.
• Removes a couple of unused deps.
• Updates the reqwest dependency in the proving service.
• Replaces once_cell with LazyLock.
• Renames miden-remote-provers to miden-proving-service-client.
• Updates the Makefile.

[SantiagoPittella]

In the original PR, there was this comment:

Question: why do we need to do take() here now? (I don't think this was needed before, right?)

This is the answer:

I replaced the .take with a .as_ref and then cloning the client.

This change was needed because of a clippy error. We didn't ran the linter earlier because it was only enabled when using the "async" feature and it was disabled in the Makefile. Now that we have clippy for no-std, the linter is throwing an error because the MutexGuard was held through an .await call.

cc @bobbinth

[bobbinth]

Looks good! Thank you! I left a couple of small comments inline. After these, we are good to merge.

[bobbinth]

I think this would need to be ../../LICENSE (since we are two levels deep from the root).

[bobbinth]

Cloning the client is pretty light-weight, right?

[SantiagoPittella]

Yes, it is composed only for the tonic client which isn't "big".

[SantiagoPittella]

@bobbinth I think this is ready to merge, but I saw that you moved it from the milestone. Should I merge it or wait until the new release is out?

[bobbinth]

@bobbinth I think this is ready to merge, but I saw that you moved it from the milestone. Should I merge it or wait until the new release is out?

Yes, let's merge. I mistakenly assigned a milestone to a PR (we usually assign milestones only to issues).