adapt workers based on whether an operation is CPU or IO-bound

Use runtime.GOMAXPROCS(0) as worker count for CPU-bound tasks,
repo.Connections() for IO-bound task and a combination if a task can be
both. Streaming packs is treated as IO-bound as adding more worker
cannot provide a speedup.

Typical IO-bound tasks are download / uploading / deleting files.
Decoding / Encoding / Verifying are usually CPU-bound. Several tasks are
a combination of both, e.g. for combined download and decode functions.
In the latter case add both limits together. As the backends have their
own concurrency limits restic still won't download more than
repo.Connections() files in parallel, but the additional workers can
decode already downloaded data in parallel.

cmd/restic/delete.go

@@ -18,8 +18,6 @@ func DeleteFilesChecked(gopts GlobalOptions, repo restic.Repository, fileList re
 	return deleteFiles(gopts, false, repo, fileList, fileType)
 }
 
-const numDeleteWorkers = 8
-
 // deleteFiles deletes the given fileList of fileType in parallel
 // if ignoreError=true, it will print a warning if there was an error, else it will abort.
 func deleteFiles(gopts GlobalOptions, ignoreError bool, repo restic.Repository, fileList restic.IDSet, fileType restic.FileType) error {
@@ -40,7 +38,9 @@ func deleteFiles(gopts GlobalOptions, ignoreError bool, repo restic.Repository,
 
 	bar := newProgressMax(!gopts.JSON && !gopts.Quiet, uint64(totalCount), "files deleted")
 	defer bar.Done()
-	for i := 0; i < numDeleteWorkers; i++ {
+	// deleting files is IO-bound
+	workerCount := repo.Connections()
+	for i := 0; i < int(workerCount); i++ {
 		wg.Go(func() error {
 			for id := range fileChan {
 				h := restic.Handle{Type: fileType, Name: id.String()}