[low-code connectors] fix so we don't display yaml when debug flag is turned off

[brianjlai]

What

For some reason I put the debug statement in the wrong place. It's poor DX to get debug statements when you don't have the --debug on.

How

We store the parsed yaml on the declarative stream. We can just print it later in the streams()method and at this point the debug/info level has been properly set.

Note: We don't need to do this for check() because check invokes the streams() method. Otherwise check would print it twice

[sherifnada]

how does moving this help? the debug level was already set above it no?

[brianjlai]

Actually when we construct the declarative source in _read_and_parse_yaml_file() in source.py (here's how its implemented for source-sendgrid), this is done before we start entrypoint.py.launch(). main.py for sengrid

So we don't actually know yet if the user specified --debug since that gets interpreted in launch(). When we move it to streams()this gets called from entrypoint.py after log level is set.

[brianjlai]

and the above debug level should be removed since its effectively a no-op, since it gets reassigned in entrypoint.

[sherifnada]

We should probably add a todo to resolve this but nbd for this pr

[brianjlai]

I removed the debug level in the above code, so I don't think we'd need any follow ups. As it stands now the flow of the read command is:

1. parse yaml
2. save it to self._source_config
3. parse command line args in entrypoint
4. assign logger level to DEBUG or INFO based on presence of --debug
5. handle read
6. emit debug message with parsed YAML if on
7. process sync