Add a comment explaining why 94-character sets have a size of 95.

Author: j4james

src/terminal/adapter/charsets.hpp

@@ -43,6 +43,11 @@ namespace Microsoft::Console::VirtualTerminal
         return rhs == lhs;
     }
 
+    // Note that the 94-character sets are deliberately defined with a size of
+    // 95 to avoid having to test the lower bound. We just alway leave the first
+    // entry - which is not meant to be mapped - as a SPACE or NBSP, which is at
+    // least visually equivalent to leaving it untranslated.
+
     typedef CharSet<L'\x20', 95> AsciiBasedCharSet;
     typedef CharSet<L'\xa0', 95> Latin1BasedCharSet94;
     typedef CharSet<L'\xa0', 96> Latin1BasedCharSet96;