Fixing entity bugs

[wcpettus]

Fixes

• scheduled_log now performs on_get in try/except loop to catch dripline errors and not crash, instead throwing a warning and returning
• calibrate now forces dict calibrations to be mapped to string, reflecting how scarab reads and passes calibration information

Testing

• tested with a few example services to ensure it works at small scale

Prior to merging for releases:

• update the project's version in the top-level CMakeLists.txt file
• update the appVersion to be the new container image tag version in chart/Chart.yaml

[nsoblath]

This feels overly broad --- all thrown exceptions are turned into warnings in the log file. There's no way to have an actual error that results in an error reply. And no information about the failure is passed back to the user.

Do we perhaps want to consider something where the on_get() function returns an optional warning/error message? This could be passed back to the user in the reply message, perhaps using a warning return code.

[wcpettus]

This is in the scheduled_log method, so by construction nothing can be passed as a reply because there is neither user initiator nor reply. This action is happening according to the scheduler (should only route through https://github.com/driplineorg/dripline-python/blob/develop/dripline/core/entity.py#L223). That was my logic for putting a very broad catch here.

I thought it better to handle it here so that the service continues running, guaranteed. Rather than messing with the on_get() method, where we could worry about intercepting what gets returned in the user reply (for example, users with dl_agent or queries that are part of a cmd would not be passing through this part of the code).

But maybe I'm missing something.