Ideas to improve Create a collection of paired datasets dialog

[bernt-matthias]

Describe the bug

Most of it comes from discussions (here).

I think Auto-pair link and docs should be more prominent (I think this is the primary way to import data).

1. The Auto-pair link itself could be a bit more prominent
2. Also in the help Auto-pair should be more prominent

- To pair datasets, you can:
- - Click a dataset in the forward-column to select it then click a dataset in the reverse-column reverse column
- - Click one of the Pair these datasets buttons in the middle column to pair the datasets in a particular row.
- - Click Auto-pair to have your datasets automatically paired based on name
+ To pair datasets, you can:
+ - Click Auto-pair to have your datasets automatically paired based on name
+ - Manually pair the datasets by:
+   1. Click a dataset in the forward-column to select it 
+   2. Click a dataset in the reverse-column reverse column
+   3. Click one of the Pair these datasets buttons in the middle column to pair the datasets in a particular row.

Make the help more visible: Until now I never realized that there is any. It's quite hidden with the little triangle.

Document better how the element identifiers are derived from the dataset names and the filters. To me it seems that part matching the filter text is just removed. But I'm just guessing from my experience with the dialog. Explicit docs would be better IMO.

Also the idea came up to use (more advanced) regular expressions to define how the name is constructed, e.g. by using named patterns (maybe like the tool xml discover_datasets syntax). But this is probably already be covered by the rule builder. Still if its easy to implement it might be more convenient for the user.

Fix:

- Note: forward slashes (\\) are not needed.
+ Note: backward slashes (\\) are not needed.`

Maybe indicate more clearly that the collection identifiers are editable.

Galaxy Version and/or server at which you observed the bug
Galaxy Version: 23.1