[to #42322933] Fix bug in UT daily

1. Fix bugs in daily test
2. Fix a bug that the updating of lr is before the first time of updating of optimizer
    TODO this will still cause warnings when GA is above 1
3. Remove the judgement of mode in text-classification's preprocessor to fit the base trainer(Bug)
     Update some regression bins to fit the preprocessor
4. Update the regression tool to let outer code modify atol and rtol
5. Add the default metric for text-classification task
6. Remove the useless ckpt conversion method in bert to avoid the requirement of tf when loading modeling_bert
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10430764

data/test/regression/sbert-base-tnews.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2bce1341f4b55d536771dad6e2b280458579f46c3216474ceb8a926022ab53d0
+size 151572

data/test/regression/sbert_nli.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44e3925c15d86d8596baeb6bd1d153d86f57b7489798b2cf988a1248e110fd62
-size 62231
+oid sha256:6af5024a26337a440c7ea2935fce84af558dd982ee97a2f027bb922cc874292b
+size 61741

data/test/regression/sbert_sen_sim.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ff17a0272752de4c88d4254b2e881f97f8ef022f03609d03ee1de0ae964368a
-size 62235
+oid sha256:bbce084781342ca7274c2e4d02ed5c5de43ba213a3b76328d5994404d6544c41
+size 61745