feat!: refactor points and add absolute coords calculation

[EscapedGibbon]

close: #469

[targos]

I didn't expect the generator to be a private method.
I thought we discussed:

• .points(kind) generator function
• .absolutePoints getter with cache
• .relativePoints getter with cache

[EscapedGibbon]

I didn't expect the generator to be a private method. I thought we discussed:

• .points(kind) generator function
• .absolutePoints getter with cache
• .relativePoints getter with cache

We decided that it would be easier that the generator would treat both cases for absolute and relative coordinates. And since the code for both getters became similar, we decided to put it into one getter with a parameter that lets the user choose what kind of coordinates one wants. The coordinates are still cached separately though.

[stropitek]

We decided that it would be easier that the generator would treat both cases for absolute and relative coordinates.

That is not contradicting what @targos said. The point is that it should not be private.

And since the code for both getters became similar, we decided to put it into one getter with a parameter that lets the user choose what kind of coordinates one wants.

It's no longer a "getter" if it takes parameters. Let's go with the initial idea: relativePoints and absolutePoints getters. Otherwise we would have to find another name to differentiate the generator from the function returning the cached points.

[stropitek]

API SGTM but the documentation explanations related to the new fields don't seem clear enough to me. We should discuss it IRL.

[EscapedGibbon]

I wrote "Images/Masks" instead of "RoiMap" for absolute coordinates because RoiMap doesn't have origin property.

[stropitek]

Can you do the relative / absolute tests with the same data? It's nice to be able to compare the result of each with one another.