proposal: drop std/io from the first stabilization targets

[kt3k]

We are still in the middle of process of migrating from Reader/Writer-based IO to Web Streams IO.

We had to undeprecate the part of std/io recently because some of IO operations cannot be replaced by Web Streams IO (ex readSync/writeSync, seek, etc). However we still aim to completely remove Reader/Writer-based IO (and therefore std/io) in the future (though it might take long time).

At this point, we can't decide on what should be/shouldn't be included in std/io. So I suggest we should drop std/io from the first stabilization targets.

[BlackAsLight]

We had to undeprecate the part of std/io recently because...

Wasn't that proposal rejected? #5032

[kt3k]

We had to undeprecate the part of std/io recently

This references these PRs: #4128 #4184 #4247 #4343

[iuioiua]

I'm for dropping @std/io from removal targets for now. But we should review and remove APIs that are worth removing. There are many APIs in @std/io that either lack sufficient use cases or overlap in functionality. Having a smaller/simpler API surface is a win for both the user and the maintainer.

[kt3k]

io has been excluded from the stable in #4600