Add all HA error numbers and descriptions to perror.

Add reminder to header.

This is already fixed in smarter ways in future codebases, and this 
codebase is unlikely to change, since new development is forbidden
here.

Bug#41112: crash in mysql_ha_close_table/get_lock_data with alter table

The problem is that the server wasn't handling robustly failures
to re-open a table during a HANDLER .. READ statement. If the
table needed to be re-opened due to it's storage engine being
altered to one that doesn't support HANDLER, a reference (dangling
pointer) to a closed table could be left in place and accessed in
later attempts to fetch from the table using the handler. Also,
if the server failed to set a error message if the re-open
failed. These problems could lead to server crashes or hangs.

The solution is to remove any references to a closed table and
to set a error if reopening a table during a HANDLER .. READ
statement fails.

Bug#41112: crash in mysql_ha_close_table/get_lock_data with alter table

The problem is that the server wasn't handling robustly failures
to re-open a table during a HANDLER .. READ statement. If the
table needed to be re-opened due to it's storage engine being
altered to one that doesn't support HANDLER, a reference (dangling
pointer) to a closed table could be left in place and accessed in
later attempts to fetch from the table using the handler. Also,
if the server failed to set a error message if the re-open
failed. These problems could lead to server crashes or hangs.

The solution is to remove any references to a closed table and
to set a error if reopening a table during a HANDLER .. READ
statement fails.

There is no test case in this change set as the test depends on
a testing feature only available on 5.1 and later.

Both of our own implementations of rint(3) were inconsistent with the
most common behavior of rint() on those platforms that have it: round
to nearest, break ties by rounding to nearest even.

Fixed by leaving just one implementation of rint() in our source tree,
and changing its behavior to match the most common native
implementations on other platforms.

slave.

In mixed mode, if we create a temporary table and do some update which switch to ROW format,
the format will keep in ROW format until the session ends or the table is dropped explicitly. 
When the session ends, the temp table is dropped automaticly at cleanup time.
but it checks only current binlog format and so skip insertion of DROP TABLE instructions into binlog.
So the temp table can't be dropped correctly at slave.

Our solution is that when closing temp tables at cleanup time we check both binlog format and binlog mode,
and we could write DROP TABLE instructions into binlog if current binlog format is ROW but in MIX mode.

                  
If secure-file-priv was set on slave, it became unable to execute
LOAD DATA INFILE statements sent from master using mixed or
statement-based replication.
                  
This patch fixes the issue by ignoring this security restriction
and checking if the files are created and read by the slave in the
--slave-load-tmpdir while executing the SQL Thread.

related to the backport of the patch for bug#38719

  
Signed integer format specifier forced to print the binlog header with server_id
negative if the unsigned value sets the sign-bit ON.
  
Fixed with correcting the specifier to correspond to typeof(server_id) == ulong.

Updated with the correct error message.

Moved the test case for the bug into a separate file (and restored the 
original innodb_mysql test setup).
Used the new wait_show_condition test macro to avoid the usage of sleep

These are only 5.0's fixes being merged.
5.1 and 6.0 Unix-removals will occur in different patches.

Re-enabling mysqlbinlog.test on Windows - removed the use of grep/sed