Heap_bloat/Btree_bloat : Lock_timeout and not statement_timeout

[Krysztophe]

The current btree_bloat (and probably others ?) have statement_timeout = 30s.

It would be cleaner to have lock_timeout too, to a much lower value (eg 3s).
If the probe fails, I'd like to know if this comes from a very heavy lock on the tables, or from the probe's duration.

PS1 : is there any way to customize the timeouts ?

PS : statement_timeout (and lock_timeout) values should appear in the error messages.