MIGRATION: Handle Jira auto-numbered ordered list syntax

[toddfarmer]

This sample issue demonstrates a gap in the jira2markdown syntax conversion utility, which apparently does not recognize Jira's auto-numbered ordered list syntax. This results in it being left as-is, which then renders as markdown headers instead. This probably needs to be fixed before migration.

[toddfarmer]

Markdown's automatic ordered list syntax is 1. rather than # . For example:

1. Test
1. Test again
1. Third test 

is rendered as:

1. Test
2. Test again
3. Third test

Because jira2markdown will also convert Jira header syntax (h1. , h2. , etc.) to markdown header syntax (# , ## , etc.), any conversion from ordered list syntax needs to be done in a manner to avoid conversion of proper headers to ordered list syntax.

[jorisvandenbossche]

Which jira2markdown implementation are you using? Is this something we could try to fix there?

[toddfarmer]

It appears that jira2markdown does support this conversion, but is confused by leading spaces before the ordered list identifiers present in the sample issue:

 # POC ...
 # The language ...

instead of:

# POC ...
# The language ...

The latter converts as expected, while the former is left untouched. I suspect the trigger is uncommon in ARROW tickets, but can be identified and fixed prior to migration relatively easily. I may also report as an issue jira2markdown. Edit: It's already reported.