Autoforge

I got a request to add a initializing function that takes the original deph of the image into account when initializing. In theory this should give you a nice background/foreground separation, but in practice it can degrade the quality of the output. In addition you need the VRAM to run the depth model. I currently don't recommend it as the output is not as good as without it, but I will leave it in for now.

The other big change is the final pruning stage that removes layers that are not needed during printing. With this and all the other pruning steps this should now give printable outputs that are usable with AMS printers and don't waste any filament.

**Full Changelog**: https://github.com/hvoss-techfak/AutoForge/compare/v1.2.1...V1.3.0

**Full Changelog**: https://github.com/hvoss-techfak/AutoForge/compare/v1.2.0...v1.2.1

The new release is finally ready.
With this release we switched from JAX to PyTorch.

Originally I used JAX mainly because I used JAX in another project and it seemed extremely fast.
I also wanted to build a project with a library I wasn't familiar with, mainly to see if I could do it.

Well, I can say that it can be done, and JAX is very fast, but also very cumbersome in some ways. Switching to PyTorch allows me (and maybe other people) to work on the library faster, as I (and probably many other people) are much more familiar with PyTorch.
The library is now about 5-10% slower, but this is hardly noticeable.

Anyway, the library is now in PyTorch.

Also, I tried a lot of different things to get a limit on colors and color swaps to work, and finally settled on a post-hoc pruning step where we remove the colors / color swaps that increase our loss function the least.

This finally solves issue 1

What's Changed
* Update args related docs by rparrett in https://github.com/hvoss-techfak/AutoForge/pull/10
* re-add random_seed for torch version by cjbnc in https://github.com/hvoss-techfak/AutoForge/pull/12

New Contributors
* rparrett made their first contribution in https://github.com/hvoss-techfak/AutoForge/pull/10

**Full Changelog**: https://github.com/hvoss-techfak/AutoForge/compare/v1.1.3...v1.2.0

I printed some of my filament, captured it with a DSLR camera, whitebalanced everything and ran a bayesian search for better color accuracy. The improvement seems to be quite substantial.

**Full Changelog**: https://github.com/hvoss-techfak/AutoForge/compare/v1.1.2...v1.1.3

First pypi release.


What's Changed
* There are a lot of changes under the hood. Giving far better and faster results.
* We are done from 8 minutes to around 2 minutes for processing an image.
* Colors are now more accurately solved.
* We now solve on a small image and transfer it to a large image at the end.
* We now support pruning to reduce the filament changes that have to be done.
* re-fit parameters for thickness to opacity function by cjbnc in https://github.com/hvoss-techfak/AutoForge/pull/4

New Contributors
* cjbnc made their first contribution in https://github.com/hvoss-techfak/AutoForge/pull/4

**Full Changelog**: https://github.com/hvoss-techfak/AutoForge/compare/v1.1.0...v1.1.2

Reworked color pipeline and some nice to have features.
This will be my final version for the foreseeable future.

**Full Changelog**: https://github.com/hvoss-techfak/AutoForge/compare/v1.0.0...v1.1.0