Overhaul the spot_detection_2d tool and add support for single-image processing

[kostrykin]

cc @rmassei

There are a few breaking changes here, mostly because I could not comprehend what the original code was doing in several cases. The new version relies on blob detection filters from skimage.feature and supports the LoG, DoG, and DoH detectors. Besides, the columns of the TSV output file have changed to lower-case, and the TSV now also includes scale and radius columns.

---

FOR THE CONTRIBUTOR — Please fill out if applicable

Please make sure you have read the CONTRIBUTING.md document (last updated: 2024/04/23).

• License permits unrestricted use (educational + commercial).

If this PR adds or updates a tool or tool collection:

• This PR adds a new tool or tool collection.
• This PR updates an existing tool or tool collection.
• Tools added/updated by this PR comply with the Naming and Annotation Conventions for Tools in the Image Community in Galaxy (or explain why they do not).

[kostrykin]

There is a really weird issue with the first test.

When running in planemo test, the test data input1.tif is loaded as an image with shape (198, 356) (which is wrong, because it should be (5, 198, 356)). When running the same script with the command line copy & pasted from the test results report, in the same Conda environment created by planemo, the image is correctly recognized as (5, 198, 356).

I'm stuck with this at the moment.
Workaround added in bc288c4

[kostrykin]

The tool produces a tabular list of detections.

This list can then be rasterized using the new version of the points2binaryimage tool: #133

After rasterization:

• Visualization can be performed using the Overlay images tool (segmentation contours over image).
• Feature extraction should also be possible using the Extract 2D features tool.

@rmassei Please feel free to review this and #133.