Combining system queries into one batch.

[OXYAMINE]

Is your feature request related to a problem? Please describe.
Very slow performance working with large MySQL databases (big number of objects) and significant network lag.
Nowadays most of us working with cloud databases where physical location could be on the other side of the planet.
Even with fast connections network lag is sufficient to affect user experience.
Especially when many SQL statements are being sent in a loop.

For example when MySQL DB connection is established Heidi sends 5 queries one by one:
SELECT CONNECTION_ID()\
SHOW STATUS\
SELECT NOW()\
SHOW VARIABLES\
SHOW DATABASES\
Consider network lag of 150-200ms between each calls and total delay is up to a second (in reality it's even more)

Next is even worse:
Opening database - 12 separate calls:
*SELECT DEFAULT_COLLATION_NAME FROM information_schema.SCHEMATA WHERE SCHEMA_NAME='c'\
SHOW TABLE STATUS FROM c\
SHOW FUNCTION STATUS WHERE Db='c'\
SHOW PROCEDURE STATUS WHERE Db='c'\
SHOW TRIGGERS FROM c\
SELECT DEFAULT_COLLATION_NAME FROM information_schema.SCHEMATA WHERE SCHEMA_NAME='information_schema'\
SHOW TABLE STATUS FROM information_schema\
SHOW FUNCTION STATUS WHERE Db='information_schema'\
SHOW PROCEDURE STATUS WHERE Db='information_schema'\
SHOW TRIGGERS FROM information_schema\
SHOW EVENTS FROM information_schema\
SELECT , EVENT_SCHEMA AS Db, EVENT_NAME AS Name FROM information_schema.EVENTS WHERE EVENT_SCHEMA='4'\

Next is even more worse: when typing "." Heidi UI tries to use autocomplete function and fetches all object names.
IT sends separate SHOW CREATE PROCEDURE or SHOW CREATE FUNCTION for each object in the database!
Imagine I have 500-700 objects in the database (sometimes more): 75 sec - 2.3 min delay is guaranteed.
Yes this is a one-off operation (in the session) and next call will be cached but this first time is a killer!
Reconnected to the database? have to wait again.

Describe the solution you'd like
I think combining queries into one batch (where possible) should remove network lag problem.

Describe alternatives you've considered
I don't see any alternatives, only if autocomplete feature could be disabled. But this is not what we want.

Additional context
N\A

[ansgarbecker]

Please disable "Full table status" in your session to see if that makes a difference:

Apart from that, some of the queries won't be quicker if combined in a batch.

Also, you should probably disable the completion proposal:

[OXYAMINE]

@ansgarbecker Get full table status doesn't help as it is mainly about procedures and functions.
Disabling "Enable autocompletion" checkbox helps but why the whole UI feature needs to be disabled if there is a better way of sending requests?
Also autocompletion isn't related to system calls at establishing connection or opening a database.