Implement blocking embedded-io traits

[tage64]

This PR implements the blocking traits from embedded-io. It refactors out the asynch module and makes the asynch traits available by a feature flag instead. Feel free to merge the PR, but it would require a breaking change.

• Remove the asynch module and put the BufferedRead and BufferedWrite structs available on crate root.
• Implement the blocking Read/Write traits from embedded-io.

[pantsman0]

@rmja +1 for this. Do you have any appetite for these changes?

I would love to use something like this for my port of the image crate for no_std+alloc envs.

[rmja]

Is the recommendation nowadays to split the crates in two (like embedded-io or embedded-io-async), or is it to separate on features?

[pantsman0]

From my experience, there is still a pretty even split. Having them together can create churn if one side has a lot more changes than the other; but conversely, having them split may mean that consumers have to rely on both depending on their in downstream users.

I think, given that you have the async traits, it would be better to have both here and enable the async versions as a feature. Publishing a buffered-io-sync would be fine but kind of undesirable as async versions are usually the ones that are tagged, and moving the current types to a buffered-io-async breaks the crate for the current users.

[rmja]

It is currently being investigated whether this should be included in embedded-io* directly. @lulf will get back.

[pantsman0]

Are you referring to rust-embedded/embedded-hal#627 ? That seems to be dead, but if there is discussion elsewhere I can jump in there

[lulf]

I didn't realize it had already been discussed. Seems like maybe splitting into buffered-io and buffered-io-async would be the best move?

[rmja]

I think I agree. Please feel free to update this PR or create a new, then I will make sure that it gets merged and published.