Remove wording that could be deemed to be perjorative (GH-9287)

Lib/_pyio.py

@@ -2283,7 +2283,7 @@ class TextIOWrapper(TextIOBase):
             # current pos.
             # Rationale: calling decoder.decode() has a large overhead
             # regardless of chunk size; we want the number of such calls to
-            # be O(1) in most situations (common decoders, non-crazy input).
+            # be O(1) in most situations (common decoders, sensible input).
             # Actually, it will be exactly 1 for fixed-size codecs (all
             # 8-bit codecs, also UTF-16 and UTF-32).
             skip_bytes = int(self._b2cratio * chars_to_skip)

Modules/_ctypes/libffi_msvc/README

@@ -372,8 +372,8 @@ single-precision anyway.  This causes one test to fail (the `many
 arguments' test).
 
 
-What's With The Crazy Comments?
-===============================
+What's With The Cryptic Comments?
+=================================
 
 You might notice a number of cryptic comments in the code, delimited
 by /*@ and @*/. These are annotations read by the program LCLint, a

Modules/_ctypes/libffi_osx/README

@@ -372,8 +372,8 @@ single-precision anyway.  This causes one test to fail (the `many
 arguments' test).
 
 
-What's With The Crazy Comments?
-===============================
+What's With The Cryptic Comments?
+=================================
 
 You might notice a number of cryptic comments in the code, delimited
 by /*@ and @*/. These are annotations read by the program LCLint, a

Tools/clinic/clinic.py

@@ -2787,7 +2787,7 @@ class str_converter(CConverter):
 
 #
 # This is the fourth or fifth rewrite of registering all the
-# crazy string converter format units.  Previous approaches hid
+# string converter format units.  Previous approaches hid
 # bugs--generally mismatches between the semantics of the format
 # unit and the arguments necessary to represent those semantics
 # properly.  Hopefully with this approach we'll get it 100% right.