doc/rst2html: some few fixes for enumerated and bullet lists

[a-mr]

First of all, fix #16041 : enumerated lists parsing is re-implemented — now there is no requirement, that it should contain some line after enumerator line, and it does not produce spurious block quotes as in #16041 . Also (1) and (2) were not parsed as enumerated list in nimdoc/rst2html/expected/rst_examples.html, most probably it was the same problem.

Also now rst.nim correctly handles alphabetic characters as enumerators:

a. string1
b. string2

while previously it turned them into 1. string1 2. string2.

According to the spec rst.nim checks that

• enumerator numbers/letters are in sequence and produces separate lists if they are not
• items belong to the same kind, and otherwise parses as 2 separate lists:

  1. list1
  a. list2
  

It can now generate auto-enumerator properly:

2. expect2
#. expect3

b. expect b
#. expect c

The first commit addresses a few blank lines problem for enumerated and bullet lists:

* this is c1
* this is c2

* this is c3


* this is c4



* this is c5

was not parsed as a single list, while Multiple successive blank lines are equivalent to a single blank line

cc @narimiran

[timotheecour]

• great, also seems to fix nim doc; maybe add a doc comment in the test file nimdoc/testproject/testproject.nim that exercises this?

• this should be documented somewhere after (or in this) PR, maybe manual.rst or doc/docgen.rst

[a-mr]

@narimiran please review.

[narimiran]

Looks good!

Since we also support markdown, the following is a valid ordered list:

1. fist
1. second
1. third

(The behaviour is the same as in your #. examples: auto-increment)

Could you add support for that too?

[Araq]

Could you add support for that too?

Can be done in a followup PR, I'm merging this now.