- lexer-strings.rb: avoid an exception on utf8 surrogate pair codepoints

[Earlopain]

Closes #855

Starting from Ruby 2.4, these are a syntax error but I don't see an easy way of representing such strings. Right now the parser actually crashses (in all versions) so I'd say it's an improvement.

Output of executing puts "\u{D800}" on all ruby versions:

Output

===================1.8===================
u{D800}
===================1.9===================
���
===================2.0===================
���
===================2.1===================
���
===================2.2===================
���
===================2.3===================
���
===================2.4===================
surrogate.rb:1: invalid Unicode codepoint
puts "\u{D800}"
         ^
===================2.5===================
surrogate.rb:1: invalid Unicode codepoint
puts "\u{D800}"
         ^~~~
===================2.6===================
surrogate.rb:1: invalid Unicode codepoint
puts "\u{D800}"
         ^~~~
===================2.7===================
surrogate.rb:1: invalid Unicode codepoint
puts "\u{D800}"
         ^~~~
===================3.0===================
surrogate.rb:1: invalid Unicode codepoint
puts "\u{D800}"
         ^~~~
===================3.1===================
surrogate.rb:1: invalid Unicode codepoint
puts "\u{D800}"
         ^~~~
===================3.2===================
surrogate.rb: --> surrogate.rb
invalid Unicode codepoint
> 1  puts "\u{D800}"
surrogate.rb:1: invalid Unicode codepoint (SyntaxError)
puts "\u{D800}"
             ^

===================3.3===================
surrogate.rb: 
surrogate.rb:1: invalid Unicode codepoint (SyntaxError)
puts "\u{D800}"
             ^

===================3.4===================
surrogate.rb: --> surrogate.rb

invalid Unicode escape sequence

> 1  puts "\u{D800}"

surrogate.rb:1: syntax error found (SyntaxError)
> 1 | puts "\u{D800}"
    |          ^~~~ invalid Unicode escape sequence
  2 | 

I used this script to check that integer.chr behaves the same on all ruby versions:

bounds = []
valid1 = true
valid2 = true
(0..(0x110000 - 1)).each do |num|
  begin
    valid1 = valid2
    num.chr(Encoding::UTF_8)
    valid2 = true
  rescue RangeError
    valid2 = false
  ensure
    bounds << num if valid1 != valid2
  end
end
puts bounds

Output

===================1.8===================
num_char.rb:7: uninitialized constant Encoding (NameError)
        from num_char.rb:4:in `each'
        from num_char.rb:4
===================1.9===================
55296
57344
===================2.0===================
55296
57344
===================2.1===================
55296
57344
===================2.2===================
55296
57344
===================2.3===================
55296
57344
===================2.4===================
55296
57344
===================2.5===================
55296
57344
===================2.6===================
55296
57344
===================2.7===================
55296
57344
===================3.0===================
55296
57344
===================3.1===================
55296
57344
===================3.2===================
55296
57344
===================3.3===================
55296
57344
===================3.4===================
55296
57344