Clarify in docs that init --batch doesn't drop old database objects

If init is run with the --batch option and a schema migration is done
old database objects (e.g. tables, columns) are not dropped
automatically. Clarify this in the documentation of the --batch option.

Change-Id: I6be33c838b18cfd789a1580ea00e60b064f3e44d
Signed-off-by: Edwin Kempin <edwin.kempin@sap.com>

Documentation/pgm-init.txt

diff --git a/Documentation/pgm-init.txt b/Documentation/pgm-init.txt
index c53c57d..ae5e471 100644
--- a/Documentation/pgm-init.txt
+++ b/Documentation/pgm-init.txt
@@ -28,6 +28,11 @@
 	Run in batch mode, skipping interactive prompts.  Reasonable
 	configuration defaults are chosen based on the whims of
 	the Gerrit developers.
++
+If during a schema migration unused objects (e.g. tables, columns)
+are detected they are *not* automatically dropped, but only a list of
+SQL statements to drop these objects is provided. To drop the unused
+objects these SQL statements have to be executed manually.
 
 \--no-auto-start::
 	Don't automatically start the daemon after initializing a