Legacy database wrapper was not supporting this properly supporting the use case where `model_id<>client_id`

See merge request !241

Doc cleanup

See merge request !240

Making the bob.bio.base database interface support samplesets where the reference_id is different from subject_id. There are some databases protocols that needs such support. Furthermore the legacy database interface was not supporting this properly

Raised a warning when data is not available to process vanilla-biometrics

See merge request !239

Created the function `get_temp_directory`

See merge request !238

[Fix] Correction of scores path

See merge request !236

[Fix] Add parenthesis else the scores filename is just ``output`` (which is a directory) when write_metadata_scores is False

[ztnorm][scorewriter] Updates

See merge request !235

[znorm][nosetests] Fixed ZNorm tests with the new scorewriter

Fixed issues in the ScoreWriter

See merge request !234

CSVSampleLoaders as transformers

See merge request !233

Move code

See merge request !232

Moved CSVBaseSampleLoader from bob.bio.base to bob.pipelines. Implemented a better version of the path_discovery function on bob.extension

Renamed CSVDatasetDevEval to CSVDataset

See merge request !231

moved VALID_DASK_CLIENT_STRINGS to bob.pipelines

See merge request !228

Check stateless

See merge request !229

# Conflicts:
#   bob/bio/base/pipelines/vanilla_biometrics/vanilla_biometrics.py