scripts.avocado: Simplify error streams detection

Avoid unnecessary complexity of quering for enabled logs and only support
os.environ to show the full traceback.

Additionally avoid problems with redirected sys.stderr by writing
directly to fd 2, which is STDERR.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

scripts/avocado

@@ -16,7 +16,6 @@
 
 import sys
 try:
-    import logging
     import os
     import tempfile
     import traceback
@@ -29,11 +28,10 @@ except:
 
 
 def handle_exception(*exc_info):
-    # Print traceback into avocado.app.traceback if initialized
+    # Print traceback if AVOCADO_LOG_DEBUG environment variable is set
     msg = "Avocado crashed:\n" + "".join(traceback.format_exception(*exc_info))
-    log = logging.getLogger("avocado.app.tracebacks")
-    if getattr(log, "handlers", False):
-        log.error(msg)
+    if os.environ.get("AVOCADO_LOG_DEBUG"):
+        os.write(2, msg + '\n')
     # Store traceback in data_dir or TMPDIR
     crash_dir = None
     prefix = "avocado-traceback-"
@@ -52,17 +50,7 @@ def handle_exception(*exc_info):
     # Print friendly message in console-like output
     msg = ("Avocado crashed unexpectedly: %s\nYou can find details in %s"
            % (exc_info[1], name))
-    for name in ("avocado.app", "avocado.test", None):
-        log = logging.getLogger(name)
-        for handler in getattr(log, "handlers", []):
-            if isinstance(handler, logging.StreamHandler):
-                break
-        else:
-            continue
-        log.critical(msg)
-        break
-    else:       # Log not found, use stderr and hope it's enabled
-        sys.stderr.write(msg + '\n')
+    os.write(2, msg + '\n')
     # This exit code is replicated from avocado/core/exit_codes.py and not
     # imported because we are dealing with import failures
     sys.exit(-1)