Note calls to other functions that need the ref instead of a slice

[llogiq]

This also applies to the needless_pass_by_value lint, though I haven't taken the time to figure out how to test it. Apart from that it will only note the calls/method calls, not the called functions/methods. Perhaps we should be able to avoid linting if the called functions are in external crates.

I'll leave both of those weaknesses to followup PRs.

[oli-obk]

Should we do this at all? I mean your example test is a false positive imo. Do you have a real use case that is not already covered by the repeated application of the other lints?

[llogiq]

I think noting it is a net win, because if someone changes the caller, but fails to change the caller, they will get a compile error. If they then complain, we can point them to the note. Also, we can extend this to omit linting if the caller is external to this crate.

[oli-obk]

I think we should just not lint sub_function. If the other functions are changed, it'll start linting anyway with the existing lints.

[llogiq]

I think it's safe to close this, as I no longer know what it should have done.