doc(inline/iter_strings_lossy): describe different behaviors

[wongjiahau]

Preface

When I copied the code from the terminal-inline example, I expected it to work as shown in the README Screenshot:

However, when I copied the code to my own project, it behaved differently. Specifically, the word was split by whitespace instead of non-alphanumeric characters.

It took me a while to figure out that the culprit was the unicode flag, which I did not enable in my project.

Reference

• similar/src/text/abstraction.rs

  Lines 68 to 74 in 47f4625

  	/// Tokenizes into unicode words.
  	#[cfg(feature = "unicode")]
  	fn tokenize_unicode_words(&self) -> Vec<&Self>;
  	/// Tokenizes into unicode graphemes.
  	#[cfg(feature = "unicode")]
  	fn tokenize_graphemes(&self) -> Vec<&Self>;