…1502)

* fix: update urls to fixed BERT models

* fix: table in pretrained_vectors docpage

* fix: change bert_config_path to .../config.json

* fix: add tokenizer_config.json, config.json to BERT-sentence models

* fix: en_core_web_sm loading error

* docs: fix comments grammar

* remove: unpopular configs and components

* refactor: squad_ru_torch_bert -> squad_ru_bert + squad_ru_bert_infer moved to torch

* refactor: squad_torch_bert[_infer] -> squad_bert[_infer]

* refactor: ner_rus_bert_torch -> ner_rus_bert

* refactor: ner_ontonotes_bert[_mult]_torch -> ner_ontonotes_bert[_mult]

* refactor: ner_conll2003_torch_bert -> ner_conll2003_bert

* insults_kaggle_bert_torch -> insults_kaggle_bert

* refactor: ranking_ubuntu_v2_torch_bert_uncased -> ranking_ubuntu_v2_bert_uncased

* Revert "refactor: squad_torch_bert[_infer] -> squad_bert[_infer]"

This reverts commit fb5bfed.

* Revert "refactor: ner_conll2003_torch_bert -> ner_conll2003_bert"

This reverts commit 7fee5e5.

* Revert "refactor: ranking_ubuntu_v2_torch_bert_uncased -> ranking_ubuntu_v2_bert_uncased"

This reverts commit e656079.

* Revert "refactor: ner_ontonotes_bert[_mult]_torch -> ner_ontonotes_bert[_mult]"

This reverts commit bb41a58.

* remove: jupyter example notebooks

* refactor: dependencies is weakened to allow run deeppavlov on python 3.8/3.9

* fix: joblib import and doc build errors

* feat: added python3 tests and lowered pytoch version upper bound

* fix: Jenkinsfile venv6 usage

* refactor: reverted minor jenkinsfile changes

* feat: docker containers initial commit

* fix: Dockerfile and cmd.sh errors

* feat: call container tests from jenkinsfile

* fix: added entrypoint to image

* feat: printenv in jenkinsfile and cmd.sh for debut purposes

* fix: tfidf_logreg_en_faq compatibility with newer scikit-learn versions

* fix: tfidf_logreg_autofaq compatibility with sklearn 0.24+

* fix: removed test_tf_layers for python3.8+

* refactor: torch.nn.modules.module.ModuleAttributeError -> AttributeError

* feat: skip tensorflow-based model tests on python 3.8/3.9

* fix: locales for hdt installation

* fix: added pybind11 to image to fix kbqa/el

* refactor: building python from sources in docker images

* fix: escape dollar sign in the Jenkinsfile

* fix: added pip upgrade to fix cryptography installation error

* fix: RECORD file not fould error for installation of the older pybind11

* fix: skip_tf_config for configs located in test_src_dir

* fix: en_ranker_pop_wiki_test

* fix: update tfidf_logreg_autofaq_psearch logreg pkl to newer sklearn version

* refactor: spacy version update

* feat: added build-essential to dockerfile and updated kenlm to fix some tests

* fix: lxml python3.9 build

* Update transformer version in bert_ranker

* delete bert_config_file from .json

* remove: transformers28.txt from requirements

Co-authored-by: Fedor Ignatov <ignatov.fedor@gmail.com>

* Remove zh squad

* Remove tests

* Fix docs

Co-authored-by: Fedor Ignatov <ignatov.fedor@gmail.com>

* Remove elmo and elmo classifiers

* Delete elmo tests

* Fix docs

* docs: returned metric and dataset name for elmo

* Add Python code sample for ELMo

* fix: docs build

* docs: removed elmo model from README.md

* docs: removed team

* docs: added 3.8 and 3.9 support info

* refactor: jinja2 requirement moved from common requirements to docs

Co-authored-by: Fedor Ignatov <ignatov.fedor@gmail.com>

* Remove ranking

* Fix config

* Remove bert_sep_ranker

* Remove bert_sep_ranker

* Fix docs

* Clean up rankers

* Restore rel_ranking

* Fix references order

* Fix docs

Co-authored-by: Fedor Ignatov <ignatov.fedor@gmail.com>