Prefer function as a parameter instead of simple trailing closure

[masters3d]

This is an exception to #54.

If the trailing closure is simple then prefer

result.map(stripTrailingWhitespace)

instead of

result.map { stripTrailingWhitespace($0) }

reference:
exercism/swift#209 (comment)

[marcelofabri]

I don't think this is directly related to #54, as it would just validate this behavior:

// triggers
result.map({ stripTrailingWhitespace($0) })

// ok
result.map { stripTrailingWhitespace($0) }

[masters3d]

"Trailing closure syntax should be used whenever possible."?

[marcelofabri]

I don't think the description totally matches what it'd be validated. Maybe @jpsim can confirm, but I feel that rule would just validate that if it's a closure and it's the last parameter, trailing syntax should be used. Function references would always be valid for that rule.

[jpsim]

@marcelofabri's last comment is spot on 👌. I would like a rule that can recommend using result.map(stripTrailingWhitespace) over result.map { stripTrailingWhitespace($0) } but in practice, that's challenging to build because you need to detect whether or not doing so would be ambiguous (e.g. cannot resolve between function overloads by the same name).