Update retention time correction step

[hechth]

Seems like the current RT alignment step doesn't use the computed clusters for the RT correction but again does some clustering or alignment based on tolerances and only uses the sample name.

There are 2 options

• read the sample name from the files so that we don't need the clustering step
• actually use the clusters for the RT correction

The rt correction step is just based on establishing matching points ("clusters") between the template and the individual samples and then a smoothed curve over the differences is created and applied to the individual samples.

Steps

• find where the clusters are computed in the retention time alignment (https://github.com/RECETOX/recetox-aplcms/blob/master/R/adjust.time.R)
• instead, read the cluster from the input file and skip the computation step
• remove the parameters that are no longer needed
• update the test cases as the results might be different

[hechth]

We should pass the sample name to the correct time function and the sample name can then be read from the parquet file metadata in the galaxy wrapper.