smoke-test for async fn with mir-opt-level=0

[RalfJung]

MIR opt levels heavily influence which MIR transformations run, and we barely test non-default opt levels. I am particularly worried about async fn lowering and how it might (not) work when the set of preceding MIR passes changes -- see #70073.

This adds some basic smoke testing, where at least a few async fnrun-pass test are ensured to also work with mir-opt-level=0.

[rust-highfive]

r? @nikomatsakis

(rust_highfive has picked a reviewer for you, use r? to override)

[jonas-schievink]

I wouldn't do this for the size tests as those could be affected by other optimizations running. For the smoke tests this seems good though.

[RalfJung]

The size can be affected by MIR optimizations?

[RalfJung]

Okay this does not touch the size tests any more.

[jonas-schievink]

The size can be affected by MIR optimizations?

Potentially yeah, at least in the future when we have some more advanced optimizations.

[RalfJung]

@jonas-schievink any chance you could review this?

[jonas-schievink]

Looks good. If you want you can also do this for some of the generator tests for some more diversity. But generally r=me

[RalfJung]

@jonas-schievink like so?

[jonas-schievink]

@bors r+

[bors]

📌 Commit 3a129df has been approved by jonas-schievink

[bors]

⌛ Testing commit 3a129df with merge d7afaa7...

[bors]

☀️ Test successful - checks-azure
Approved by: jonas-schievink
Pushing d7afaa7 to master...