From https://docs.perl6.org/routine/contains, `contains` coerces its invocant to a string and searches for a substring starting from a certain position (index 0 by default):

    # Let's drop `<>` (quote-words constructor) and 
    # use a more familiar array of strings:
      
    my @menu = 'hamburger', 'fries', 'milkshake';
      
    # Let's look at the definition of the `Str` method in 
    # the `List` class from which the `Array` class inherits:
    
    #`{
      method Str(List:D: --> Str:D)
      Stringifies the elements of the list and joins
      them with spaces (same as .join(' ')).
    }

    # Thus, after being stringfied, @menu is treated as 
    # the string 'hamburger fries milkshake'
    
    # With this is in hand, we can gauge the possible
    # result of the following statements:
    
    say @menu.contains('hamburger');            # True 
    say @menu.contains('hot dog');              # False 
    say @menu.contains('milk');                 # True
    say @menu.contains('er fr');                # True 
    say @menu.contains(<es mi>);                # True

If you'd like instead to search the array for words as elements, we could use Perl 6's set functions. However, I don't know if they're as performant as regular string searches:

    say 'hamburger' (<=) @menu;  # True 
    say 'fries' (<=) @menu;      # True 
    say 'hot dog' ∈ @menu;       # False 
    say 'milk' ∈ @menu;          # False 
    say 'er fr' ∈ @menu;         # False 

`∈` is the unicode equivalent of `(<=)`.

---

I'm not sure if I'm missing something in the "odd design choices" in this specific case so it'd be great if you could elaborate a little bit further here.