更新

(?<=                        # Match something preceded by
  (?<![\s\S])               #          at the start of the file
  #match\n                  # '#match'
  [\s\S]*                   # and anything in between:
)                           # 
^(# .+)\n                   # A heading followed by
([\s\S]+?)                  # its corresponding content, which might be as long as possible,
(?=\n# |<!--ID|(?![\s\S]))  # until another heading, '<!--ID' or the end of file.

试试on regex101.com。
由于ECMAScript不支持\A和\Z（尽管有a proposal可以添加它们），我们必须使用负查找来做：(?![\s\S]) / (?<![\s\S])。[\s\S]匹配any single character，所以(?![\s\S])只能在找不到后续字符的位置匹配：字符串的结尾。同样的解释也适用于(?<![\s\S])。

原始答案

(?:\A#match|\G(?!\A))  # Match '#match' at the start of the file, or the end of the last match
\s*                    # followed by 0 or more whitespaces;
\K                     # we forfeit everything we just matched
^(# .+)\n              # then match and capture a heading, before continuing to the next line
([\s\S]+?)             # and capture the section's content,
(?=\n# |\Z)            # which must precedes either another heading or the end of file.

试试on regex101.com。
这利用了以下元序列：

• \A：整个字符串的最开始
• \G：最后一场比赛的结束或\A
• \G(?!\A)：仅最后一场比赛的结束
• \K：放弃与其左边表达式匹配的所有内容
• \Z：字符串的结尾或最后一行结束符之前的位置 iff 它也是整个字符串的最后一个字符。