Rémy Coutable 创作于 3 years ago ，Lin Jen-Shin 提交于 3 years ago

Signed-off-by: Rémy Coutable <remy@rymai.me>

This adds a RuboCop rule that disallows the usage of
`SomeModel.establish_connection`, as this pattern can break database
configurations and slows down the tests.

See https://gitlab.com/gitlab-org/gitlab/-/issues/338653 for more
information.

Auto-corrected with Rubocop

If a spec such as `wiki_pipeline_spec.rb` included a NULL byte in the
example text, then the `insert-rspec-profiling-data` script would fail
with an `invalid byte sequence` error
(e.g. https://gitlab.com/gitlab-org/gitlab/-/jobs/386527522).

To avoid this, we do two things:

1. Drop the NULL byte from the example test.
2. Modify the `insert-rspec-profiling-data` script to sanitize the CSV.

Closes https://gitlab.com/gitlab-org/gitlab/issues/120012

Changes all calls to data_source_exists? to table_exists?
since that is the intent of these calls

Changes all calls to data_source_exists? to table_exists?
since that is the intent of these calls

We also enable the rubocop that makes it mandatory

Instead of inserting a row after each example to an external database,
we save the CI profiling reports into the `rspec_profiling` directory
and insert the data in the update-tests-metadata CI stage.  This should
make each spec run faster and also reduce the number of PostgreSQL
connections needed by concurrent CI builds.

`scripts/insert-rspec-profiling-data` also inserts one file at a time
via the PostgreSQL COPY command for faster inserts. The one side effect
is that the `created_at` and `updated_at` timestamps aren't available
since they aren't generated in the CSV.

Closes https://gitlab.com/gitlab-org/gitlab-ee/issues/10154