Allow supplying custom workflow yaml for infer_with_model job

[fm3]

URL of deployed dev instance (used for testing):

• https://___.webknossos.xyz

Steps to test:

• set up local worker, with branch from https://github.com/scalableminds/voxelytics/pull/3605
• yarn enable-jobs
• train a model
• infer with the model, both with and without custom yaml
• example custom yaml: https://github.com/scalableminds/voxelytics/pull/3605#issuecomment-2250234257

TODOs:

• Backend: add this param to the route, to job_args
• Worker:
  • use the passed yaml https://github.com/scalableminds/voxelytics/pull/3605
  • decide about the --until call, compare discussion
  • fill out template part of passed yaml
• Frontend:
  • Show textarea for yaml, append it to route params
  • Prefill textarea

Issues:

• contributes to [AI models] AI model training and management #7670

---

• Removed dev-only changes like prints and application.conf edits
• Considered common edge cases

[philippotto]

probably read for review now, right?

[fm3]

probably read for review now, right?

The worker side isn’t ready, I’ll do some end-to-end testing next. I think that more checkboxes may arise (e.g. we will likely drop the option to select the output layer name)

[fm3]

slipped in here: the decimals were lost due to integer division

[fm3]

@MichaelBuessemeyer could you please take a look at Philpp’s frontend comments?

[philippotto]

testing went ok 👍

[MichaelBuessemeyer]

@MichaelBuessemeyer could you please take a look at Philpp’s frontend comments?

Thanks for the ping. Done :)

[philippotto]

there's a conflict to resolve, but other than that, everything seems good 👍