Count Matching Files Recursively

The monitoring team wants a tiny helper they can point at any directory.

Task

Create the executable script /home/student/bin/count-files.sh that takes one argument, a directory, and:

• prints only a number: how many regular files whose name ends in .out exist anywhere under that directory (subdirectories included)
• counts files with spaces or newlines in their names correctly, and ignores directories named *.out
• if the argument is not an existing directory, prints not a directory: <arg> to stderr and exits with code 1

The grader runs it against hidden directory trees.