Cross Correlation Popup (located in Display / Verify)

[indecisiveuser]

This PR creates a cross-correlation popup to test image registration quality. As current image quality tools in multiview-reconstruction are interest point-based, this would provide an independent check.

The functionality is as follows: 1. provides the option to compute metrics on all tiles (default) or selected tiles
2. allows the user to select the downsampling and channel(s)
3. calculates the cross correlation for the chosen overlapping tile pairs
4. displays the results in a table sorted by correlation in ascending order. The average intensity of each ViewID is also displayed.

Note: this PR is meant to work in conjunction with this BigStitcher PR

[StephanPreibisch]

TODO: use https://github.com/JaneliaSciComp/BigStitcher-Spark?tab=readme-ov-file#examples first dataset, stitch before and after the updates

[indecisiveuser]

I calculated the stitching on the above dataset (downsampled 2,2,1) with both the new version and mvrecon 7.0.3.

All r values are identical across versions.
There's a slight discrepancy in the actual coordinates of each tile (see the below table), but it's practically negligible and can be explained by the fact that the GlobalOpt.java solver changed from serial to concurrent.

setupId	Δtx	Δty	Δtz	3D distance
0	0.0000	+0.0001	0.0000	0.0001
1	0.0000	0.0000	0.0000	0.0000
2	0.0000	+0.0002	0.0000	0.0002
3	+0.0001	+0.0002	+0.0001	0.00025
4	+0.0001	+0.0002	0.0000	0.00022
5	+0.0001	+0.0003	+0.0001	0.00033
6	+0.0001	+0.0003	0.0000	0.00032
7	+0.0001	+0.0003	+0.0001	0.00033
8	+0.0001	+0.0003	0.0000	0.00032
9	+0.0001	+0.0003	0.0000	0.00032
10	+0.0001	+0.0003	+0.0001	0.00033
11	+0.0001	+0.0002	+0.0001	0.00025
12	+0.0002	+0.0003	+0.0001	0.00037
13	+0.0001	+0.0004	+0.0001	0.00042
14	+0.0002	+0.0003	+0.0001	0.00037
15	+0.0002	+0.0003	+0.0001	0.00037
16	+0.0001	+0.0003	+0.0001	0.00033
17	+0.0001	+0.0003	0.0000	0.00032

[StephanPreibisch]

great, thanks so much!

[StephanPreibisch]

let's merge once we looked also at the BigStitcher one here JaneliaSciComp/BigStitcher#162 - this one does not compile because it needs a new release of multiview-reconstruction I guess?

[indecisiveuser]

Sure, sounds good. I just fixed the compilation error on the BigStitcher one; the pom was using a snapshot version. I needed to use a snapshot to test out the functionality but it should be fine once they're both merged.