chore: Add installation instructions and fix dependencies

[nikolasrieble]

The core change here are (1) installation instructions to the README.
Following the instructions myself, the initial command

uv run docling-eval --help

failed with

❯ uv run docling-eval --help
2025-11-27 18:41:43,161 - INFO - PyTorch version 2.8.0 available.
Traceback (most recent call last):
  File "/Users/niko.rieble/code/docling-eval/.venv/bin/docling-eval", line 4, in <module>
    from docling_eval.cli.main import app
  File "/Users/niko.rieble/code/docling-eval/docling_eval/cli/main.py", line 115, in <module>
    from docling_eval.prediction_providers.aws_prediction_provider import (
        AWSTextractPredictionProvider,
    )
  File "/Users/niko.rieble/code/docling-eval/docling_eval/prediction_providers/aws_prediction_provider.py", line 7, in <module>
    import boto3
ModuleNotFoundError: No module named 'boto3'

uv groups should be optional. If something is needed for a basic --help command, it should be a base dependency. To resolve this, I (2) move two dependencies from the hyperscalar group into the base group.

And finally, (3) with uv sync we update the uv lockfile.

This PR closes #168

[github-actions]

✅ DCO Check Passed

Thanks @nikolasrieble, all your commits are properly signed off. 🎉

[mergify]

Merge Protections

Your pull request matches the following merge protections and will not be merged until they are valid.

🟢 Enforce conventional commit

Wonderful, this rule succeeded.

Make sure that we follow https://www.conventionalcommits.org/en/v1.0.0/

• title ~= ^(fix|feat|docs|style|refactor|perf|test|build|ci|chore|revert)(?:\(.+\))?(!)?:

[cau-git]

@nikolasrieble Thanks for making an effort here. This repo definitely lacks documentation on the installation process.

I would make certain changes though:

1. We do actually want the hyperscaler dependencies to remain optional, so the better path would be to fix the code to not require them by default.
2. The installation instructions you created are for a development setup. We should add first the instructions to install docling-eval straight from pypi.

Would you be up to iterate on this?

[nikolasrieble]

@nikolasrieble Thanks for making an effort here. This repo definitely lacks documentation on the installation process.

I would make certain changes though:

1. We do actually want the hyperscaler dependencies to remain optional, so the better path would be to fix the code to not require them by default.
2. The installation instructions you created are for a development setup. We should add first the instructions to install docling-eval straight from pypi.

Would you be up to iterate on this?

Fully agree with (1). My naive proposal would be: Lets merge this and then iterate, i.e. do another PR to add (2), and fix (1) in yet another.

Often making smaller steps ends up being faster.