Revert `tests/testdata` changes and address flake8 issues

tests/testdata/wmt20-en-zh-v1-res.json

-{"results": {"wmt20-en-zh": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.00014170297316825535, "chrf_stderr": 6.590669847391838e-05, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-en-zh": 1}}
+{"results": {"wmt20-en-zh": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.00014170297316825535, "chrf_stderr": 6.590669847391838e-05, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-en-zh": 1}}
\ No newline at end of file

tests/testdata/wmt20-fr-de-v0-greedy_until

-8a4b65c59dcac6591d46261909ee92ebcf41c19ee7442b12842302b2d8aeb36f
+8a4b65c59dcac6591d46261909ee92ebcf41c19ee7442b12842302b2d8aeb36f
\ No newline at end of file

tests/testdata/wmt20-fr-de-v0-res.json

-{"results": {"wmt20-fr-de": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.01143193767396364, "chrf_stderr": 0.00012555271954563658, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-fr-de": 0}}
+{"results": {"wmt20-fr-de": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.01143193767396364, "chrf_stderr": 0.00012555271954563658, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-fr-de": 0}}
\ No newline at end of file

tests/testdata/wmt20-iu-en-v0-greedy_until

-97bf664a8efa54b5366b8341f77b418106dd0cb26169d5b2d0144e4d3d2bc5c9
+97bf664a8efa54b5366b8341f77b418106dd0cb26169d5b2d0144e4d3d2bc5c9
\ No newline at end of file

tests/testdata/wmt20-iu-en-v0-res.json

-{"results": {"wmt20-iu-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.012204628007572778, "chrf_stderr": 8.944407532175802e-05, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-iu-en": 0}}
+{"results": {"wmt20-iu-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.012204628007572778, "chrf_stderr": 8.944407532175802e-05, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-iu-en": 0}}
\ No newline at end of file

tests/testdata/wmt20-ja-en-v0-greedy_until

-1fd846f3c0104e794eb380dae7f648592092ab8bf59234c26d0a671bbbc28df1
+1fd846f3c0104e794eb380dae7f648592092ab8bf59234c26d0a671bbbc28df1
\ No newline at end of file

tests/testdata/wmt20-ja-en-v0-res.json

-{"results": {"wmt20-ja-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.010703148854351403, "chrf_stderr": 0.00022242113108130186, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-ja-en": 0}}
+{"results": {"wmt20-ja-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.010703148854351403, "chrf_stderr": 0.00022242113108130186, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-ja-en": 0}}
\ No newline at end of file

tests/testdata/wmt20-km-en-v0-greedy_until

-fb4ec81bb89c70df7e21b43e0e882915b7b71a2a85bb8d4b59e0c7938baaa4c2
+fb4ec81bb89c70df7e21b43e0e882915b7b71a2a85bb8d4b59e0c7938baaa4c2
\ No newline at end of file

tests/testdata/wmt20-km-en-v0-res.json

-{"results": {"wmt20-km-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.015142474534585969, "chrf_stderr": 0.0001518735048829897, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-km-en": 0}}
+{"results": {"wmt20-km-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.015142474534585969, "chrf_stderr": 0.0001518735048829897, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-km-en": 0}}
\ No newline at end of file

tests/testdata/wmt20-pl-en-v0-greedy_until

-89274499d84176b1ffe4eaec06f2c89ca807342384dc946c2e348d00116aaade
+89274499d84176b1ffe4eaec06f2c89ca807342384dc946c2e348d00116aaade
\ No newline at end of file

tests/testdata/wmt20-pl-en-v0-res.json

-{"results": {"wmt20-pl-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.01353367757716276, "chrf_stderr": 0.00018386199249976465, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-pl-en": 0}}
+{"results": {"wmt20-pl-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.01353367757716276, "chrf_stderr": 0.00018386199249976465, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-pl-en": 0}}
\ No newline at end of file

tests/testdata/wmt20-ps-en-v0-greedy_until

-c3976465e3709b4bc371175cc1494c69fe096ea4ba7d114da779d2baa0a47466
+c3976465e3709b4bc371175cc1494c69fe096ea4ba7d114da779d2baa0a47466
\ No newline at end of file

tests/testdata/wmt20-ps-en-v0-res.json

-{"results": {"wmt20-ps-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.015192865365105723, "chrf_stderr": 0.00011334541381539086, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-ps-en": 0}}
+{"results": {"wmt20-ps-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.015192865365105723, "chrf_stderr": 0.00011334541381539086, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-ps-en": 0}}
\ No newline at end of file

tests/testdata/wmt20-ru-en-v0-greedy_until

-1477ab6542c26bd0222cc1aded174f33bf8d04d1cf6a1c0959aeca4ff3779adc
+1477ab6542c26bd0222cc1aded174f33bf8d04d1cf6a1c0959aeca4ff3779adc
\ No newline at end of file

tests/testdata/wmt20-ru-en-v0-res.json

-{"results": {"wmt20-ru-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.013344639906399232, "chrf_stderr": 7.583552652374546e-05, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-ru-en": 0}}
+{"results": {"wmt20-ru-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.013344639906399232, "chrf_stderr": 7.583552652374546e-05, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-ru-en": 0}}
\ No newline at end of file

tests/testdata/wmt20-ta-en-v0-greedy_until

-111ea3efdc08f1cf536631b9426c3a20e482c575d009d2a8c71f59c027578eec
+111ea3efdc08f1cf536631b9426c3a20e482c575d009d2a8c71f59c027578eec
\ No newline at end of file

tests/testdata/wmt20-ta-en-v0-res.json

-{"results": {"wmt20-ta-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.013841110664859798, "chrf_stderr": 0.00018476696850880766, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-ta-en": 0}}
+{"results": {"wmt20-ta-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.013841110664859798, "chrf_stderr": 0.00018476696850880766, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-ta-en": 0}}
\ No newline at end of file

tests/testdata/wmt20-zh-en-v0-greedy_until

-07dbadfd6f2b2b9462ab6187dbfaabae6e5192ab89a8e4ede9237834b9364dd1
+07dbadfd6f2b2b9462ab6187dbfaabae6e5192ab89a8e4ede9237834b9364dd1
\ No newline at end of file

tests/testdata/wmt20-zh-en-v0-res.json

-{"results": {"wmt20-zh-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.008438201290981157, "chrf_stderr": 0.0001109053964076822, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-zh-en": 0}}
+{"results": {"wmt20-zh-en": {"bleu": 0.0, "bleu_stderr": 0.0, "chrf": 0.008438201290981157, "chrf_stderr": 0.0001109053964076822, "ter": 1.0, "ter_stderr": 0.0}}, "versions": {"wmt20-zh-en": 0}}
\ No newline at end of file

tests/testdata/wnli-v0-loglikelihood

-2ffd304d6096416eb29607e2e7642b1d6043163624967bcf4c4fc00fddc6c721
+2ffd304d6096416eb29607e2e7642b1d6043163624967bcf4c4fc00fddc6c721
\ No newline at end of file