feat(ListGroup): improve configuration options

[MaxTru]

• Place items on bottom if sorting is disabled
• Make it possible to disable autoOpening via new parameter autoOpen

related to #96

[barmac]

It does not only change the configuration, but also the behaviour when you add new items. Note that without the proposed changes, the newly added items are stacked one on another. This means that if I add two items, the first one of these takes position no. 2. Check out this footage:

old-listener.mp4

With this PR, only the newly added item (singular) is placed on top while the item added previously is placed as in the "sorted" order. So if add two items, the former one can be placed anywhere within the "old" items. Check out the footage:

listener.mp4

So if we merge this, it will behave differently than the nested list.

[barmac]

Check out the comment above.

[MaxTru]

It does not only change the configuration, but also the behaviour when you add new items. Note that without the proposed changes, the newly added items are stacked one on another. This means that if I add two items, the first one of these takes position no. 2. Check out this footage:

old-listener.mp4
With this PR, only the newly added item (singular) is placed on top while the item added previously is placed as in the "sorted" order. So if add two items, the former one can be placed anywhere within the "old" items. Check out the footage:

listener.mp4
So if we merge this, it will behave differently than the nested list.

So you mean it should only sort in case we close / open again?

[barmac]

At least that's how it is implemented in nested list and the current version of list group.

[andreasgeier]

So you mean it should only sort in case we close / open again?

Yes. Sort only kicks in on close/open or when switching element selection. The requirement behind this is that list items must not jump when adding or editing items.

[MaxTru]

Thanks @andreasgeier and @barmac for the hint and explanation. With the support of @pinussilvestrus this was fixed:

[barmac]

It works as expected. I created a follow-up issue for an a11y issue I noticed: #107. It is not a regression.

[andreasgeier]

✅   LGTM