fix(rosetta): diagnostics from infused snippets were not ignored

[kaizencc]

Recently we implemented the infused metadata that turns strict=false
and loose=true for the specific snippet. However we still output the
diagnostics along with other snippets (with isError=true). One thing we
could do is make these diagnostics isError=false, but I argue that we
want to ignore these diagnostics entirely. They aren't actually diagnostics;
we will find them in the cache and they do compile.

---

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

[rix0rrr]

I agree with the conclusion, but I'm not sure the premise is right. There is a bug, but I think it lies somewhere else:

Recently we implemented the infused metadata that turns strict=false and loose=true for the specific snippet. However we still output the diagnostics along with other snippets (with isError=true).

AFAIK setting strict=false should prevent isError from turning true, no? Why is that not working?

I argue that we want to ignore these diagnostics entirely. They aren't actually diagnostics; we will find them in the cache and they do compile.

The diagnostics must originate from trying to compile the snippet. So my counter-question to this observation would be: if we find them in the cache, then why are we trying to compile them in the first place?

[mergify]

Thank you for contributing! ❤️ I will now look into making sure the PR is up-to-date, then proceed to try and merge it!

[mergify]

Merging (with squash)...

[mergify]

Merging (with squash)...