This reverts commit 7287c4b4.

revert last commit, cascading ApplicationModules should be avoided instead (and should be prevented in future)

Cache accessors obtained through TestFacility::getScalar() and TestFacility::getArray() by the application test code, to avoid the loss of the current value if a decorated accessor is obtained again.

added support for the std::string data type and throw an exception if the type given by the user is unknown

also make sure for DeviceModules to always return the same sub-module with the [] operator when passing the same sub-module name

fixed not returning the same submodule on a second call of the ControlSystemModule::operator[] with the same module name

allow connecting a consuming node to a consuming node, if one of them already has been connected to a feeder.

don't crash the application if Profiler::startMeasurement() is called while the measurement is already running, instead just ignore the second call

Renamed Application::testableModeThreadName() into Application::threadName(), since this will be used for other purposes as well.

moved the xml generator main function into a new header file, to avoid defining main() multiple times by mistake.

moved group read/write and []/() oprerators from the VariableGroup class into the Module class, to make the functionality also available in ApplicationModules

added function to obtain the qualified name of the variable, including the hierarchy path. not yet used.

- use this return type in testable mode to detect when a queue overflow happened (testable mode counter should not increase in that case)

When optimising the connections (i.e. merging networks), the trigger receivers for triggered networks are now removed from the trigger network. This was previously leading to crashes.

renamed VariableNetwork::addTriggerReceiver() into VariableNetwork::addNodeToTrigger() for better clarity

renamed VariableNetworkNode getTriggerReceiver() into VariableNetworkNode getNodeToTrigger() for better clarity

use the new readLatest(), which has been added to the latest DeviceAccess version, to implement the naturally expected behaviour when using a push-type variable with a poll-type accessor

When attaching a trigger to a node, the resulting triggered node is stored and later returned when the same trigger is attached again to the same node. This fixes crashes when joining two networks sharing the same node triggered by the same trigger, if the triggering node was technically different.

Instead of making the DeviceModule return the same VariableNetworkNode for the same register, an optimisation step before realising the connections (but after all connections have been defined) is performed. In this step, all networks sharing the same (device-type) feeder will be merged. This is more flexible and avoids problems with (potentially different) triggers etc.

A DeviceModule now returns the same VariableNetworkNode when asked for the same register multiple times. This leads to a new restriction that registers can now only be either read or written. This restriction can be mitigated by using the logical name mapper to create an alias of the register. This has been implemented for the tests and the demo application.

- moved the sleep in Application::testableModeLock() so it helps better against false positive stall detections
- added comments to the growingly complex code of Application::testableModeLock()

preparation for future changes: The TestDecoratorRegisterAccessor now gets a ProcessArray as argument, not the more generic NDRegisterAccessor. This allows it to use specific features of the ProcessArray.

detect stalled tests and print a list of variables which contain data not processed by the application

- prevent spamming of some messages in debug testable mode