[Feature] Forward reference for datasets/datalinks

[swampie]

Currently seqerakit does support dataset creation but to have an e2e flow where a pipeline input can reference a Seqera Platform's dataset url is necessary to use a mix of python and yaml.

It would be great to have the possibility to define in the same yaml a forward reference for entities created at different stage of the execution:

Pseudo-code

dataset:
   name: 'input.csv'
   file: 'samplesheet/rnaseq.csv'
   reference: 'my-input'
...
pipeline:
  name: rnaseq
...
launch:
   pipeline:
     name: rnaseq
     params:
        input: $ref.my-input <- that's references the url of the previously created dataset

The behavioud could also be enforced with validation where the $ref keyword (or similar) can be parsed to check if in the timeline of the execution an entity with that name has been created.

Once available, the same can be used for data-links

[ejseqera]

I've proposed a solution for this, would you be able to kindly test it out and let me know what you think? @swampie

I am trying to avoid having to specify additional key-value pairs and use the dataset name created as a reference for retrieving the dataset URL to be used as the input value in launching, and adding a pipeline.