GSVA: Update futurize support on GSVA

Hi,

I have deprecated the functions `gsvaRanks()` and `gsvaScores()`, which have been replaced by `gsvaColRanks()` and `gsvaColScores()`, in the current devel branch of GSVA (version >= 2.7.5). Because they are part of the API supported by futurize, I would like to suggest adding support for `gsvaColRanks()` and `gsvaColScores()` in futurize.

Additionally, a new function called `gsvaRowNorm()`, which has been added to this devel branch, would need also support from futurize. Section 4.4 in the single-cell [vignette](https://bioconductor.org/packages/3.24/bioc/vignettes/GSVA/inst/doc/GSVA_scRNAseq.html#calculate-gsva-scores) illustrates the use of these three functions.

Finally, in this devel branch I have also added a new support for using GSVA across independent nodes in an HPC environment, with two new functions `gsvaMap()` and `gsvaReduce()` that implement a map-reduce layer, where `gsvaMap()` can take a `BiocParallel::BatchtoolsParam` argument. The previous functions `gsvaRowNorm()`, `gsvaColRanks()` and `gsvaColScores()` take two parameters called `first` and `last` that allow the software to restrict calculations to a range of rows or columns, enabling the distribution of calculations across independent nodes and `gsvaMap()` takes care of that distribution. The manual pages of `gsvaMap()` and `gsvaReduce()` contain full details on how to use them and toy examples, and I have added a small [vignette](https://bioconductor.org/packages/3.24/bioc/vignettes/GSVA/inst/doc/GSVA_HPC.html) illustrating how to use them with the SLURM workload manager. I guess here we could add HPC support using future.batchtools somehow, let me know if I can assist somehow with this, and do not hesitate to let me know any suggestions you may have.

Thanks!