I am not sure whether anyone on this support board is an expert on OpenSwath search results.
You can build a spectral library from multiple search result files. What happens if you use the .osw files from each of your 47 runs to build a spectral library? Is there some sort of false discovery rate stuff that you are missing out on by not having a merged.osw file?
BiblioSpec looks at the filename extension in order to decide what to do with peptide search result files. If you think these .oswm files might be similar enough to a .osw file that BiblioSpec might be expecting, you could try renaming them to .osw and see whether it works. (although "model" does not sound like the name of a file that would contain a list of results).
The first version of BiblioSpec that could read .osw files was in 2018. Has OpenSwath changed the format of their output files in a way that we need to support? You can certainly send us your files and we can take a look.
You can upload large files here:
https://skyline.ms/files.url
-- Nick