Prevent unintentional reconnections to the master server

If a master_failure_mode was set to error_on_write, a reconnection to the
old master would happen after the following events:

- Master server fails and the connection is closed
- The master server recovers
- A slave fails and the connection is closed
- A replacement for the slave is searched

If these events took place, the master would be taken back into use with
an inconsistent session state.

server/modules/include/readwritesplit.h

@@ -54,7 +54,7 @@ typedef enum bref_state
     BREF_WAITING_RESULT   = 0x02, /*< for session commands only */
     BREF_QUERY_ACTIVE     = 0x04, /*< for other queries */
     BREF_CLOSED           = 0x08,
-    BREF_SESCMD_FAILED    = 0x10 /*< Backend references that should be dropped */
+    BREF_FATAL_FAILURE    = 0x10 /*< Backend references that should be dropped */
 } bref_state_t;
 
 #define BREF_IS_NOT_USED(s)         ((s)->bref_state & ~BREF_IN_USE)
@@ -62,7 +62,7 @@ typedef enum bref_state
 #define BREF_IS_WAITING_RESULT(s)   ((s)->bref_num_result_wait > 0)
 #define BREF_IS_QUERY_ACTIVE(s)     ((s)->bref_state & BREF_QUERY_ACTIVE)
 #define BREF_IS_CLOSED(s)           ((s)->bref_state & BREF_CLOSED)
-#define BREF_HAS_FAILED(s)          ((s)->bref_state & BREF_SESCMD_FAILED)
+#define BREF_HAS_FAILED(s)          ((s)->bref_state & BREF_FATAL_FAILURE)
 
 typedef enum backend_type_t
 {