patch 9.0.2064: cannot use buffer-number for errorformat

Problem:  cannot use buffer-number for errorformat
Solution: add support for parsing a buffer number using '%b' in
          'errorformat'

closes: #13419

Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Yegappan Lakshmanan <yegappan@yahoo.com>

runtime/doc/quickfix.txt

@@ -1422,6 +1422,7 @@ rest is ignored.  Items can only be 1023 bytes long.
 Basic items
 
 	%f		file name (finds a string)
+	%b		buffer number (finds a number)
 	%o		module name (finds a string)
 	%l		line number (finds a number)
 	%e		end line number (finds a number)
@@ -1461,6 +1462,11 @@ On MS-Windows a leading "C:" will be included in "%f", even when using "%f:".
 This means that a file name which is a single alphabetical letter will not be
 detected.
 
+The "%b" conversion is used to parse a buffer number.  This is useful for
+referring to lines in a scratch buffer or a buffer with no name.  If a buffer
+with the matching number doesn't exist, then that line is used as a non-error
+line.
+
 The "%p" conversion is normally followed by a "^".  It's used for compilers
 that output a line like: >
 	    ^

runtime/doc/todo.txt

@@ -5945,8 +5945,6 @@ Various improvements:
 7   Instead of filtering errors with a shell script it should be possible to
     do this with Vim script.  A function that filters the raw text that comes
     from the 'makeprg'?
--   Add %b to 'errorformat': buffer number. (Yegappan Lakshmanan / Suresh
-    Govindachar)
 7   Allow a window not to have a statusline.  Makes it possible to use a
     window as a buffer-tab selection.
 8   Allow non-active windows to have a different statusline. (Yakov Lerner)