Adding config creation features (#138)

* improve efficiency, modularization of create_main_config function

* bug-fix updating min_common_af based on user MAF threshold input

* change dtype to float

* fix list operation

* update handle_pretrained_models function logic for readability & efficiency

* issue 131 fixed - allow update of base_configurations via input_config

* Update docs for configuration file udpates

* merge in deterministic flag , optional input keys, and clobber from main

* fixup! Format Python code with psf/black pull_request

* move logging setup command

* update base-config default MAF,
add optional 'sample_files' config key,
add resolve for pretrained_model_path

* Add pytests for configuration file generation

* fixup! Format Python code with psf/black pull_request

---------

Co-authored-by: PMBio <PMBio@users.noreply.github.com>

Author: meyerkm

deeprvat/deeprvat/base_configurations.yaml

@@ -73,7 +73,7 @@ training_data:
 association_testing_data:
     dataset_config:
         min_common_af:
-            MAF: 0.01
+            MAF: 0.001
         use_common_variants: False
         use_rare_variants: True
         rare_embedding: