* gather up all the flaky test stuff into a directory
* create Jenkins Pipeline DSL for the report generation and the flaky re-testing
* have the nightly per-branch job consume the results of flaky reporting
Signed-off-by: Mike Drob <mdrob@apache.org>
Conflicts:
dev-support/Dockerfile
dev-support/Jenkinsfile
dev-support/flaky-tests/flaky-dashboard-template.html
dev-support/flaky-tests/report-flakies.py
For branches 1, includes a backport of the current version of report-flakies.py
and supporting files.