keep new line for task description (#2116)

* add keep trailing newline

* apply ruff-format

* add prompt unit test

* increment the version of tasks that have description with whitespace

* remove white spaces of leaderboard bbh

* update MMLU expected versions in output

* CI run does display the expected version=1 for mmlu subtasks, fix expected test output again

---------
Co-authored-by: haileyschoelkopf <hailey@eleuther.ai>

lm_eval/tasks/aclue/_aclue.yaml

@@ -23,4 +23,4 @@ aggregate_metric_list:
     aggregation: mean
     weight_by_size: true
 metadata:
-  version: 0.0
+  version: 1.0

lm_eval/tasks/aclue/_default_template_yaml

@@ -15,4 +15,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  version: 0.0
+  version: 1.0

lm_eval/tasks/aexams/_aexams.yaml

@@ -13,4 +13,4 @@ aggregate_metric_list:
     aggregation: mean
     weight_by_size: true
 metadata:
-  version: 0.0
+  version: 1.0

lm_eval/tasks/aexams/_default_template_yaml

@@ -15,4 +15,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  version: 0.0
+  version: 1.0

lm_eval/tasks/bbh/cot_fewshot/_bbh.yaml

@@ -33,4 +33,4 @@ aggregate_metric_list:
     weight_by_size: true
     filter_list: get-answer
 metadata:
-  version: 2.0
+  version: 3.0

lm_eval/tasks/bbh/cot_fewshot/_bbh_cot_fewshot.yaml

@@ -33,4 +33,4 @@ aggregate_metric_list:
     weight_by_size: true
     filter_list: get-answer
 metadata:
-  version: 2.0
+  version: 3.0

lm_eval/tasks/bbh/cot_fewshot/_cot_fewshot_template_yaml

@@ -24,4 +24,4 @@ filter_list:
       - function: "take_first"
 num_fewshot: 3
 metadata:
-  version: 2.0
+  version: 3.0

lm_eval/tasks/bbh/cot_zeroshot/_bbh_cot_zeroshot.yaml

@@ -33,4 +33,4 @@ aggregate_metric_list:
     weight_by_size: true
     filter_list: flexible-extract
 metadata:
-  version: 2.0
+  version: 3.0

lm_eval/tasks/bbh/cot_zeroshot/_cot_zeroshot_template_yaml

@@ -23,4 +23,4 @@ generation_kwargs:
   temperature: 0.0
 num_fewshot: 0
 metadata:
-  version: 2.0
+  version: 3.0

lm_eval/tasks/bbh/fewshot/_bbh_fewshot.yaml

@@ -32,4 +32,4 @@ aggregate_metric_list:
     aggregation: mean
     weight_by_size: true
 metadata:
-  version: 2.0
+  version: 3.0

lm_eval/tasks/bbh/fewshot/_fewshot_template_yaml

@@ -17,4 +17,4 @@ generation_kwargs:
   temperature: 0.0
 num_fewshot: 3
 metadata:
-  version: 1.0
+  version: 2.0

lm_eval/tasks/bbh/zeroshot/_bbh_zeroshot.yaml

@@ -33,4 +33,4 @@ aggregate_metric_list:
     weight_by_size: true
     filter_list: flexible-extract
 metadata:
-  version: 2.0
+  version: 3.0

lm_eval/tasks/bbh/zeroshot/_zeroshot_template_yaml

@@ -23,4 +23,4 @@ generation_kwargs:
   temperature: 0.0
 num_fewshot: 0
 metadata:
-  version: 2.0
+  version: 3.0

lm_eval/tasks/ceval/_ceval-valid.yaml

@@ -7,7 +7,7 @@ aggregate_metric_list:
   weight_by_size: true
 group: ceval-valid
 metadata:
-  version: 1.0
+  version: 2.0
 task:
   - ceval-valid_computer_network
   - ceval-valid_operating_system

lm_eval/tasks/ceval/_default_ceval_yaml

@@ -15,4 +15,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  version: 1.0
+  version: 2.0

lm_eval/tasks/cmmlu/_cmmlu.yaml

@@ -75,4 +75,4 @@ aggregate_metric_list:
     metric: acc_norm
     weight_by_size: true
 metadata:
-  version: 0.0
+  version: 1.0

lm_eval/tasks/cmmlu/_default_template_yaml

@@ -15,4 +15,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  version: 0.0
+  version: 1.0

lm_eval/tasks/gpqa/cot_n_shot/_gpqa_cot_n_shot_yaml

@@ -35,4 +35,4 @@ metric_list:
     ignore_case: true
     ignore_punctuation: true
 metadata:
-  version: 1.0
+  version: 2.0

lm_eval/tasks/gpqa/generative/_gpqa_generative_n_shot_yaml

@@ -36,4 +36,4 @@ metric_list:
     ignore_case: true
     ignore_punctuation: true
 metadata:
-  version: 1.0
+  version: 2.0

lm_eval/tasks/gpqa/n_shot/_gpqa_n_shot_yaml

@@ -18,4 +18,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  version: 1.0
+  version: 2.0