Sometimes several long running modules in a row might have caused false alarms. So seen in tests of the llrfctrl server with multiple devices (which require some time to open and initialise).

Simply add timeout to acquiring the testable mode lock and print an error message with the last thread which successfully obtained the thread. Finally throw an exception to help debugging.

Without them, the references become invalid in move operations which leads to subsequent segfaults.

* optimization for StatusAggregator
make getPriority faster by making prio map static and reduce evaluation frequency

* get rid of compiler warning

* comment for statusPrio

Co-authored-by: Dietrich Rothe <dietrich.rothe@desy.de>

Slashes are not allowed in module names, but the DeviceModule crates a status module using the CDD (or alias) as a name. The slashes are now replaced by underscores.

ApplicationCore 02.07 is the planned feature upgrade (PreriodicTrigger) for DeviceAccess 02.08, while the master already is on DeviceAccess 03.00

Did not increase the patch because of binary incompatibilities.

See #7425

It can be static and it may make sense to use it outside modules. The implementation should probably move to a Application member function to make it easier to reach from non-module code.

ScriptedInitialisationHandler: catch boost exceptions and throw ChimeraTK logic_error with improved what() message

to check whether stepApplication() can execute...

Co-authored-by: Martin Hierholzer <martin.hierholer@desy.de>