Bump dependencies invalidly assuming memory layout of SocketAddr

[faern]

Bumps net2, socket2 and miow.
Helps unblock #78802

Done as separate PR since frequent lockfile collisions is a thing... And since the main PR can't be merged until large parts of the ecosystem uses the newer crates only, so we have to start somewhere.

[rust-highfive]

r? @Mark-Simulacrum

(rust-highfive has picked a reviewer for you, use r? to override)

[Mark-Simulacrum]

@bors r+ rollup=never

[bors]

📌 Commit fccdef8762ad692b29b6f632049fbdffdce3d0e0 has been approved by Mark-Simulacrum

[bors]

⌛ Testing commit fccdef8762ad692b29b6f632049fbdffdce3d0e0 with merge 0d27e4d1ab28f7c507fd380b31823939aa703fe9...

[bors]

💔 Test failed - checks-actions

[faern]

I'm not sure I understand the CI failure. It fails at the "Mark the job as a failure" task. Not sure what went wrong or if this PR caused it or if it's just sporadic.

[Mark-Simulacrum]

Illumos builder failed; https://github.com/rust-lang-ci/rust/runs/1467496350#step:24:11749 -- bors doesn't link to the real cause, you have to scroll to find a red checkmark.

[faern]

Ah. Yeah ok. Looks like socket2 0.3.17 added a bunch of code relying on TCP_MAXSEG. This does not exist in libc for illumos yet. Or it was just merged, but not published: rust-lang/libc#1984

So I resorted to downgrading to socket2 0.3.16 for now. It's still strictly an upgrade in this PR, but just not to the very latest version.

[Mark-Simulacrum]

Please squash commits. If you'd like to note the problem with 0.3.17 in the commit message that would be fine, but I don't think we need a separate commit.

[faern]

@Mark-Simulacrum Done

[Mark-Simulacrum]

@bors r+

[bors]

📌 Commit f9220c3 has been approved by Mark-Simulacrum

[bors]

⌛ Testing commit f9220c3 with merge 88b8197...

[bors]

☀️ Test successful - checks-actions
Approved by: Mark-Simulacrum
Pushing 88b8197 to master...