remove ambiguous operator

[trittsv]

This PR removes a ambiguous operator.

It solves the following situation.

dds::core::InstanceHandle handle = domainParticipant.instance_handle();
std::ostringstream s;
s << handle;

Error:

error C2593: 'operator <<' is ambiguous 
D:\conan\short\f72057\1\include\ddscxx\dds/core/detail/TInstanceHandleImpl.hpp(105,22): message : could be 'std::ostream &operator <<( 
std::ostream &,const dds::core::TInstanceHandle<org::eclipse::cyclonedds::core::InstanceHandleDelegate> &)' [found using argument-depe 
ndent lookup]
D:\conan\short\f72057\1\include\ddscxx\dds/core/detail/TInstanceHandleImpl.hpp(112,22): message : or       'std::ostream &operator <<( 
std::ostream &,const dds::core::TInstanceHandle<org::eclipse::cyclonedds::core::InstanceHandleDelegate>)' [found using argument-depend 
ent lookup]
 message : while trying to match the argument list '(std::ostringstream, dds::core::InstanceHandle)'

[eboasson]

I wonder the vague (to me) errors in https://dev.azure.com/eclipse-cyclonedds/cyclonedds-cxx/_build/results?buildId=6872&view=logs&jobId=1600059a-7091-5d41-ccb9-416f6972ece0&j=1600059a-7091-5d41-ccb9-416f6972ece0&t=7ad8b019-d95f-5e19-9aeb-cd9e72c6a999 are caused by this. A quick scan of the log doesn't show it complaining about instance handles, but perhaps that is behind another layer of unwrapping?

If it is just the test code that suffers from this, then let's fix the test code :)

[eboasson]

@trittsv that failure seems to have something to do with an interaction between gtest and old gcc. I have no idea what/why/how/..., but it builds fine with this commit on top of yours: eboasson@ee6d57b

What do you think?

[trittsv]

@eboasson i like this solution, i will integrate it into this PR. May its a bug in the compiler, buts thats just a wild guess.