In terminate(), call interrupt only when wait_for_new_data is set

This partly reverts d5ab1447, where the existing doReadTransfer overrides
were removed. This commit gets the existing implmentations back, but
changes the function name to doReadTransferSynchronously.

Remove functions that do not exist anymore

Throws logic_error if these modifiers are used in the top level

ExceptionHandlingDecorator: doReadTransferNonBlocking() and doReadTransferLatest() must always return true because there will be new data. The recovery in postRead() takes care of it.

ExceptionHandlingDecorator: With the current (=old) the owner has to be informed about data invalidity as soon as the exception is reported, and then again after recovery.

improved places where DeviceModule changes testableMode_deviceInitialisationCounter. Now matches with deviceHasError.

Now the writing in the DeviceModule is working.

Fixes #159

- removed wrong waitForRecovery()
- some code improvements
- fixed wrong logic in doReadTransferAsync

Revert "ExceptionHandlingDecorator: re-introduced while loop. Behaviour is now closer to previous one because it re-tries the read until it succeeds"

Looping the postAction makes no sense. The loop only worked for the transfer itself.

This reverts commit f7f35074.

ExceptionHandlingDecorator: re-introduced while loop. Behaviour is now closer to previous one because it re-tries the read until it succeeds

* doGenericPostAction is void
* removed endless loop in doGenericPostAction
* only wait for recovery if application is running in multi-threaded mode
* fixed return value for doReadTransferXxx in case transfer is not allowed

Fixes #151

	TODO doGenericPostAction
	DONE all the rest

	does not compile, due to missing overrides

This case was missed in #129.

allows to control whether to wait for device recoveries in stepApplication