Implement client side retry

[germa89]

After so much research on the infamous MAPDL gRPC issues #3313 I have seen that there is a pattern in some issues:

• MAPDL server connection terminated when using parameters #2989
• PyMapdl crahes/looses connection, when running in a loop #3156
• gRPC server exceptions #2479
• test_xpl.py::test_save fails on v24.2 #2904 (Although I'm not very sure about this one)

They all presents the grpc_status:14 in the debug_error_string. Full error:

debug_error_string = "UNKNOWN:Error received from peer {created_time:"2024-06-06T11:36:14.151348247+00:00", grpc_status:14, grpc_message:"Connection reset"}"

After researching about this grpc_status:14, I have found these references:

• stack overflow question
• gRPC Status response codes
• stack overflow question 2

Hence I'm going to implement a retry mechanism, as close as possible to the gRPC interface. So we try to redo the call when this fail with this error message. I haven't decided the implementation yet.

I also believe this PR should help to fix #2479 because I'm using that model to fix it.