examples: add mnist fgm demo which makes use of RESTAPI v1

[jtsextonMITRE]

Closes #596, except that the internals will need to be reworked to use the v1 client once it is fully finished.

[jayeew]

It seems not work.
when I run this code:
experiment_id, train_ep, queue_id = upload_experiment(client, 'src/train.yml','train','training a classifier on MNIST', PLUGIN_FILES, QUEUE_NAME, QUEUE_DESC, EXPERIMENT_NAME, EXPERIMENT_DESC)
The error code 400 returned.
2024-09-27 16:23:08 [error ] Error code 400 returned. data={'group': 1, 'name': 'train', 'description': 'training a classifier on MNIST', 'taskGraph': ' dataset:\n load_dataset:\n ep_seed: $seed\n data_dir: $training_dir\n subsets: [training, validation, testing]\n image_size: $image_size\n validation_split: $validation_split\n batch_size: $batch_size \n\n model:\n create_model:\n dataset: $dataset.training\n model_architecture: $model_architecture\n input_shape: $image_size\n learning_rate: $learning_rate\n optimizer: $optimizer_name\n metrics_list:\n - name: CategoricalAccuracy\n parameters: { name: accuracy }\n - name: Precision\n parameters: { name: precision }\n - name: Recall\n parameters: { name: recall }\n - name: AUC\n parameters: { name: auc }\n dependencies:\n - dataset\n\n trained_model:\n train:\n estimator: $model\n x: $dataset.training\n callbacks_list:\n - name: EarlyStopping\n parameters:\n monitor: val_loss\n min_delta: .01\n patience: 5\n restore_best_weights: true\n fit_kwargs:\n nb_epochs: $epochs\n validation_data: $dataset.validation\n verbose: 2\n dependencies:\n - model\n\n metrics:\n compute_metrics:\n classifier: $trained_model\n dataset: $dataset.testing\n dependencies:\n - trained_model\n\n save:\n save_artifacts_and_models:\n models:\n - name: $register_model_name\n model: $trained_model\n ', 'parameters': [{'name': 'seed', 'defaultValue': '-1', 'parameterType': 'integer'}, {'name': 'optimizer_name', 'defaultValue': 'Adam', 'parameterType': 'string'}, {'name': 'learning_rate', 'defaultValue': '0.001', 'parameterType': 'float'}, {'name': 'training_dir', 'defaultValue': '/dioptra/data/Mnist/training', 'parameterType': 'string'}, {'name': 'testing_dir', 'defaultValue': '/dioptra/data/Mnist/testing', 'parameterType': 'string'}, {'name': 'image_size', 'defaultValue': '[28, 28, 1]', 'parameterType': 'list'}, {'name': 'validation_split', 'defaultValue': '0.2', 'parameterType': 'float'}, {'name': 'batch_size', 'defaultValue': '32', 'parameterType': 'integer'}, {'name': 'model_architecture', 'defaultValue': 'le_net', 'parameterType': 'string'}, {'name': 'epochs', 'defaultValue': '30', 'parameterType': 'integer'}, {'name': 'register_model_name', 'defaultValue': 'mnist_classifier', 'parameterType': 'string'}], 'queues': [30], 'plugins': [31]} method=POST response={"schema_errors": {"parameters": {"0": {"parameterType": ["Must be one of: string, float, path, uri."]}, "5": {"parameterType": ["Must be one of: string, float, path, uri."]}, "7": {"parameterType": ["Must be one of: string, float, path, uri."]}, "9": {"parameterType": ["Must be one of: string, float, path, uri."]}}}} url=http://localhost:80/api/v1/entrypoints/

Is it the parameterType error??