Fix #651. Add title manager to ensure unique displayed titles.

[cmeyer]

How I tested:

Use Processing > Generate to create 2+ data items with same name. Take FFT of results to generate new data items. Run acquisition and take multiple snapshots and see that they have unique titles.

[KRLango]

I have concerns about this PR.

What is the problem it is trying to solve?
Having unique names - but why do users want unique names, and does this solve their actual need?

During my testing, I got 2 identical suffixes #96 with 4 snapshots. While I accept this will be rare - how often do we want to accept collisions happening?

Will the random nature of the suffixes confuse users? For example, the first few I got all appeared to be just numbers #58, #96, #84. Are users going to wonder what is going on with those numbers?

I believe the implementation done for the Acquisitions is a much cleaner user experience, and I think something similar would work better here.

[cmeyer]

This may not be needed in light of the acquisition index change and it may indeed be confusing to users. We can discuss that before merging (or not merging).

However, I'm concerned that even if we decided it was a useful feature, it shouldn't ever have duplicate suffixes. So I'm wondering how that occurred. I tested by changing the minimum length to 1 and then creating multiple snapshots and when it got to a duplicated suffix out of the 0-9A-F set, it increased the length to 2. Did this happen when the '96' appeared for you, e.g. increase to length 3 to distinguish between the two '96's?

[KRLango]

I have gone back to the project I used for this. I did some snapshots using ctrl + s and some using the menu. When I found the 2 96's I think it was one with each since the names are slightly different (Display Snapshot vs Snapshot).

[cmeyer]

Ok that makes sense - but it does highlight a problem with this scheme. I'm going to change this to draft while I think it through.