In profiling integration tests, I found a couple places where per-test
overhead could be reduced:
* Avoiding disk IO by synchronizing instead of deleting & copying test
Git repository data. This saves ~100ms per test on my machine
* When flushing queues in `PrintCurrentTest`, invoke `FlushWithContext`
in a parallel.
---------
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Currently 'userxx' and 'orgxx' are both used as username in test files
when the user type is org, which is confusing. This PR replaces all
'userxx' with 'orgxx' when the user type is org(`user.type==1`).
Some non-trivial changes
1. Rename `user3` dir to `org3` in `tests/git-repositories-meta`
2. Change `end` in `issue reference` because 'org3' is one char shorter
than 'user3'
3. Change the search result number of `user/repo2` because
`user3/repo21` can't be searched now
4. Change the first org name getting from API because the result is
ordered by alphabet asc and now `org 17` is before `org25`
Other modifications are just find all and replace all.
Unit tests with SQLite are all passed.
---------
Co-authored-by: caicandong <1290147055@qq.com>
