L1processing

[tmieslinger]

This PR adds the possibility to create level1 files from level0 via the ASPEN software that is run in a docker image. That way, it is not a prerequisite to process L0 files via ASPEN or BatchASPEN manually. If L1 files exist, they are used. If not, they will be created using the docker image ghcr.io/atmdrops/aspenqc generated and described in the repo aspenqc.

The implementation includes adding the L0 to L1 processing in the pipeline dict and moving the filter_no_launch_detect function upfront. Also, several error messages are removed as (1) the launch_detect function was moved and (2) if the a-file does not exist, python will raise an AttributeError anyway.

Closes #25.

[Geet-George]

The way this is set up looks so nice! :) Thanks so much.

One question: Currently for the package to work it would have to be connected to the internet to get the docker image right? Would it be possible that we add pull in the docker image as a dependency for the package to work? This means that in-flight processing would also be possible.

[tmieslinger]

That's a really good point. So if you run it once, it will be cached on your laptop and you can use the docker image offline, too. But I'll check whether and how it could be integrated into the first installation of halodrops 👍

[tmieslinger]

Following Tobi's (@d70-t) advice, I'd start an issue on the offline processing of L0 to L1 as it's supposedly a feature creep to pull a docker image within a pip install halodrops. I would instead add a note in the docu or readme that if you plan to process L0 to L1 the first time offline, then you'd have to pull the docker image manually first.

[Geet-George]

Yeah, makes sense. Also, best to merge this asap, I guess. The :::warning::: in docu should be enough for users to know that it should be pulled once before.