input files in bescape MuSic

Current input for bescape.deconv.deconvolute_gep()

```
# Important to specify ABSOLUTE directory paths
wd = os.getcwd()
annot = wd + '/datasets/bescape/gep'
inpt = wd + '/datasets/bescape/input'
output = wd + '/datasets/bescape/output'

print(output)
# deconvolute using MuSiC - sc based basis vector
deconv.deconvolute_gep(dir_annot= annot,
dir_input= inpt
```
can we point to the actual annotation and input files rather than the input folder here? so input=/path/bulk.csv and annot=.path/eSet.rds ?

`(Please note that when single file is expected either in the scRNASeq or bulkRNASeq, it will just grab the first alphabetical file in the respective folder)`
1) if there's more than one file, the wrapper should at least filter for .csv searching for the input and .rds searching for the annotation file ? I believe it does literally just list the files independent of format and picks the first one

Start at bescape.deconv.deconvolute_gep() and the deconv.deconvolute_gep() wrapper shown in the issue. Trace how the scRNA-seq annotation and bulk RNA-seq input folders are listed, then check how direct file paths and extensions such as .csv and .rds should be handled. Done means callers can provide individual files and multiple files do not cause an unrelated first alphabetical file to be selected.

Written by the indexing model from the issue text.