added NeMoLogger unit tests

[dorotat-nv]

Adding hotfix and unit tests uncovering issue with WandbLogger in https://nvidia.slack.com/archives/C074Z808N05/p1733418209959769

DETAILS

An interesting observation: this genuinely appears to be a wandb bug, since it can only be captured once it’s converted into a Lightning hook during training. Only one out of 4 new unit tests fails when probably all of them should

The error is only uncovered only after NeMoLogger setup as n https://github.com/NVIDIA/NeMo/blob/bb895bc4b28ba99d707cb907c4496297a2a7b533/nemo/collections/llm/api.py#L852C22-L856C6
and after plt.Trainer setup at the start of the training in
https://github.com/Lightning-AI/pytorch-lightning/blob/de7c28ae865b5c9fd3ff21debebb994605f7f420/src/lightning/pytorch/trainer/trainer.py#L944
which is executed in
https://github.com/Lightning-AI/pytorch-lightning/blob/caa9e1e59436913e365bf52eeb2b07e3bf67efac/src/lightning/pytorch/trainer/call.py#L94C1-L97C34

TESTING ON BROKEN MAIN BRANCH

self = Settings(allow_offline_artifacts=True, allow_val_change=False, anonymous=None, api_key=None, azure_account_url_to_acce...andb', sync_symlink_latest='/workspace/bionemo2/wandb/latest-run', timespec='', wandb_dir='/workspace/bionemo2/wandb/')
name = 'root_dir', value = PosixPath('/tmp/pytest-of-dorotat/pytest-20/test_nemo_logger_initilized0/unit-test-loggers')

    def __setattr__(self, name: str, value: Any) -> None:
        if name in self.__class_vars__:
            raise AttributeError(
                f'{name!r} is a ClassVar of `{self.__class__.__name__}` and cannot be set on an instance. '
                f'If you want to set a value on the class, use `{self.__class__.__name__}.{name} = value`.'
            )
        elif not _fields.is_valid_field_name(name):
            if self.__pydantic_private__ is None or name not in self.__private_attributes__:
                _object_setattr(self, name, value)
            else:
                attribute = self.__private_attributes__[name]
                if hasattr(attribute, '__set__'):
                    attribute.__set__(self, value)  # type: ignore
                else:
                    self.__pydantic_private__[name] = value
            return
    
        self._check_frozen(name, value)
    
        attr = getattr(self.__class__, name, None)
        if isinstance(attr, property):
            attr.__set__(self, value)
        elif self.model_config.get('validate_assignment', None):
>           self.__pydantic_validator__.validate_assignment(self, name, value)
E           pydantic_core._pydantic_core.ValidationError: 1 validation error for Settings
E           root_dir
E             Input should be a valid string [type=string_type, input_value=PosixPath('/tmp/pytest-of...zed0/unit-test-loggers'), input_type=PosixPath]
E               For further information visit https://errors.pydantic.dev/2.9/v/string_type

/usr/local/lib/python3.10/dist-packages/pydantic/main.py:881: ValidationError

[dorotat-nv]

\build-ci

[dorotat-nv]

/build-ci

[dorotat-nv]

/build-ci

[dorotat-nv]

/build-ci

[dorotat-nv]

/build-ci

[skothenhill-nv]

/build-ci

[dorotat-nv]

/build-ci

[dorotat-nv]

/build-ci

[dorotat-nv]

/build-ci

[dorotat-nv]

/build-ci

[dorotat-nv]

/build-ci

[dorotat-nv]

/build-ci

[dorotat-nv]

/build-ci