Preserve Jira IDs from the source system in the target system when migrating

[tjaehnel]

Since Jira automatically assigns ticket IDs in the order of their creation, it is not easily possible to perserve the IDs from the source Jira project in the target Jira project.

The only solution is to add them in exactly the same order as they have been added to the source system. Since ask-jira creates EPICs and Subtasks on demand, this would break this order.

This pull request introduces two configuration options NO_AUTO_CREATE_EPICS and NO_AUTO_CREATE_SUBTASKS. Setting both to "true" disables automatic creation of issues and thus issues are added to the target system in the same order they come from the filter.

For details and known limitation, see the README change.

I've performed one Jira project migration so far and it successfully preserved the IDs as expected.

[mrts]

Hm, this is a bit more controversial... let it simmer a bit. I can see the value of this, but it is inherently fragile IMHO.

[tjaehnel]

I totally agree.
It was mainly an attempt because I really wanted to keep the IDs in the new system and it did what I logically expected.

I tried to document all pitfalls and limitation I could think of and made it optional.
In fact what this change mostly does is to (optionally) disable features ask-jira already has. Namely automatic creation of EPICs and Subtasks.
This in itself shouldn't hard too much.

What it adds is the capability to find existing EPIC and Subtask relationships and tries to rebuild them in the target system. This turns around the original idea of ask-jira and will probably work in most cases but not all.

What it makes fragile is the fact that this relies on the consecutive numbering in both the source and target system.

Maybe we should make the fragility more explicit in the readme and put some words of warning to the front.