feat(dataProcess): Add parallel processing functionality to dataProcess

[tonywu1999]

Motivation and Context

In theory, MSstats should produce more meaningful results if clients set up experiments with more bioreplicates. However, as clients begin to input larger datasets into MSstats, they have been running into issues regarding speed & memory. This pull request aims to speed up the MSstats dataProcess function by adding a parameter called numberOfCores that should allow users to perform data processing (specifically the summarization step) in parallel per protein.

This is an extension of the parallel processing PR for groupComparison

Changes

• Added new numberOfCores parameter to dataProcess function. Default is 1 core, which goes through the regular for loop approach with a progress bar indicator.
• Used the parallel package's parLapply function to process MSstatsSummarizeSingleTmp or MSstatsSummarizeSingleLinear in parallel per protein.

Testing

• Tested speed on this 2.3 GB dataset and encountered a ~30% increase in performance from 1 core to 2 cores on my M1 mac that has 8 cores (i.e. 102 seconds -> 78 seconds).
• Added unit tests for dataProcess verifying that parallel processed function emitted the same number of rows as dataProcess with default parameters.
• Callout: For some reason, input_split = split(input, input$PROTEIN) creates a variable that is listed as 35GB in R despite this dataset being 2.3GB. If this variable is created prior to exporting cluster variables, the parallel processing stalls since the input_split object takes up a huge amount of space in memory. As a result, I've refactored the multi-core approach to not create this object and instead represent the split using a list of indices instead (see protein_indices variable).

Checklist Before Requesting a Review

• [x] I have read the MSstats contributing guidelines
• [x] My changes generate no new warnings
• [x] Any dependent changes have been merged and published in downstream modules

[mstaniak]

Are there any new imports or were all these functions used in groupComparison update already? I will review this very soon

[tonywu1999]

@mstaniak

All these functions were used in the groupComparison update already.

[mstaniak]

Then the changes are OK, just two names requires an update, well done