Fix Python scripts for Windows: UTF-8 encoding (datacarpentry#485)

To avoid problems with various symbols, we have to specify the encoding
when we read files.
The actual codec name is `utf_8` but aliases like `utf8`, `utf-8`, etc
are accepted. Here, I'm using `utf-8` alias.
https://docs.python.org/3.8/library/codecs.html#standard-encodings

This fixes `make lesson-check` when running under 'Git for Windows' for
lessons that have non-cp1252 characters.

bin/lesson_check.py

@@ -215,7 +215,7 @@ def read_references(reporter, ref_path):
     result = {}
     urls_seen = set()
 
-    with open(ref_path, 'r') as reader:
+    with open(ref_path, 'r', encoding='utf-8') as reader:
         for (num, line) in enumerate(reader, 1):
 
             if P_INTERNAL_INCLUDE_LINK.search(line): continue

bin/util.py

@@ -105,7 +105,7 @@ def read_markdown(parser, path):
     """
 
     # Split and extract YAML (if present).
-    with open(path, 'r') as reader:
+    with open(path, 'r', encoding='utf-8') as reader:
         body = reader.read()
     metadata_raw, metadata_yaml, body = split_metadata(path, body)
 
@@ -160,7 +160,7 @@ def load_yaml(filename):
     """
 
     try:
-        with open(filename, 'r') as reader:
+        with open(filename, 'r', encoding='utf-8') as reader:
             return yaml.load(reader, Loader=yaml.SafeLoader)
     except (yaml.YAMLError, IOError) as e:
         print('Unable to load YAML file {0}:\n{1}'.format(

bin/workshop_check.py

@@ -408,7 +408,7 @@ def main():
     reporter = Reporter()
     check_config(reporter, config_file)
     check_unwanted_files(root_dir, reporter)
-    with open(index_file) as reader:
+    with open(index_file, encoding='utf-8') as reader:
         data = reader.read()
         check_file(reporter, index_file, data)
     reporter.report()