chore: Bump pip-compile-multi

[john-bodley]

SUMMARY

Rumor has it that the pip-compile-multi ecosystem was likely borked, probably due to changes which i) used an improper version of pip-compile-multi/pip-tools, or ii) updates to setup.py etc. which didn't update the pinned requirements.

This PR updates pip-compile-multi and updates the requirements/*.txt files without upgrading any of the packages, via:

pip-compile-multi --only-name integration --upgrade-package pip-compile-multi 
python3.8 -m pip install -r requirements/integration.txt 
pip-compile-multi --no-upgrade

Note we likely should look into something like http://dependabot.com/ or similar for ensuring dependencies remain fresh. See the Slack thread in the #committers channel here for more detail.

BEFORE/AFTER SCREENSHOTS OR ANIMATED GIF

TEST PLAN

ADDITIONAL INFORMATION

• Has associated issue:
• Changes UI
• Includes DB Migration (follow approval process in SIP-59)
  • Migration is atomic, supports rollback & is backwards-compatible
  • Confirm DB migration upgrade and downgrade tested
  • Runtime estimates and downtime expectations provided
• Introduces new feature or API
• Removes existing feature or API

[codecov]

Codecov Report

Merging #14633 (3db127a) into master (2bd0b62) will increase coverage by 0.08%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master   #14633      +/-   ##
==========================================
+ Coverage   77.39%   77.47%   +0.08%     
==========================================
  Files         958      958              
  Lines       48486    48486              
  Branches     5679     5679              
==========================================
+ Hits        37524    37563      +39     
+ Misses      10762    10723      -39     
  Partials      200      200              

Flag	Coverage Δ
hive	80.96% <ø> (ø)
mysql	81.24% <ø> (ø)
postgres	81.26% <ø> (ø)
presto	80.95% <ø> (?)
python	81.78% <ø> (+0.15%)	⬆️
sqlite	80.88% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
superset/models/core.py	89.13% <0.00%> (+0.27%)	⬆️
superset/connectors/sqla/models.py	90.26% <0.00%> (+1.44%)	⬆️
superset/db_engine_specs/presto.py	89.89% <0.00%> (+5.47%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2bd0b62...3db127a. Read the comment docs.

[villebro]

LGTM - I'd really like to look into the dependabot option to replace pip-compile-multi. If help is needed, I can try to find time to work on that.

[john-bodley]

BTW I'm a fan of pip-compile-multi as it groups dependencies by roles in a maintainable way. Additionally per the Slack thread it's important that we also update the pre-commit hooks (ideally at the same time) and I'm unsure if dependabot handles this.

@villebro if you or @nytai have time to explore this that would be great. I think one learning of auto-updating dependencies on a regular basis for a number of internal projects is it really only works if one has good test coverage, linting etc. which boosts one's confidence of the legitimacy of the PR, i.e., increases the likelihood of merging with little to no manual testing. The general lack of pylint coverage could be problematic.