fix: new column UUID conflicts in dual write

[eschutho]

Follow up to #20351 where I added some extra checks and lookups for when a column lookup returned None on dual write. We're still seeing incidents of the same error: psycopg2.errors.UniqueViolation: duplicate key value violates unique constraint "sl_columns_uuid_key" intermittently and I was able to repro again in a test when adding in an additional query on the session for existing columns. It seems that the query is triggering an autoflush which is attempting to create a new column when one already exists. I added a no-flush flag onto any newColumn queries and also made sure to pull in the id of any existing columns when updating it and adding it to the session so that SqlAlchemy will treat it as an update instead of a write.

TESTING INSTRUCTIONS

More unit tests, because we don't have a reliable way to repro this issue. But regression testing is necessary. Most issues are happening on import when the dataset already exists and is being overwritten, but for regression, thorough testing of editing and importing datasets that either do or do not exist would be needed.

ADDITIONAL INFORMATION

• Has associated issue:
• Required feature flags:
• 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 #20460 (e9e6ea6) into master (f3b289d) will decrease coverage by 0.10%.
The diff coverage is 86.95%.

❗ Current head e9e6ea6 differs from pull request most recent head 04ce15c. Consider uploading reports for the commit 04ce15c to get more accurate results

@@            Coverage Diff             @@
##           master   #20460      +/-   ##
==========================================
- Coverage   66.67%   66.56%   -0.11%     
==========================================
  Files        1739     1739              
  Lines       65111    65127      +16     
  Branches     6896     6896              
==========================================
- Hits        43414    43355      -59     
- Misses      19948    20023      +75     
  Partials     1749     1749              

Flag	Coverage Δ
hive	?
mysql	82.33% <86.95%> (-0.01%)	⬇️
postgres	82.40% <86.95%> (-0.01%)	⬇️
python	82.48% <86.95%> (-0.24%)	⬇️
sqlite	82.19% <86.95%> (-0.01%)	⬇️

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

Impacted Files	Coverage Δ
superset/connectors/sqla/models.py	89.17% <86.95%> (-0.30%)	⬇️
superset/db_engines/hive.py	0.00% <0.00%> (-85.19%)	⬇️
superset/db_engine_specs/hive.py	70.22% <0.00%> (-15.65%)	⬇️
superset/db_engine_specs/presto.py	83.64% <0.00%> (-0.63%)	⬇️
superset/db_engine_specs/base.py	88.00% <0.00%> (-0.34%)	⬇️
superset/utils/core.py	90.03% <0.00%> (-0.12%)	⬇️

---

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 f3b289d...04ce15c. Read the comment docs.

[eschutho]

@jinghua-qa If you're free, this could use some basic regression testing around dataset creation and importing. Thank you!

[jinghua-qa]

/testenv up

[github-actions]

@jinghua-qa Ephemeral environment spinning up at http://35.88.160.132:8080. Credentials are admin/admin. Please allow several minutes for bootstrapping and startup.

[jinghua-qa]

Tested regression for import and create dataset in ephermal env, LGTM!

[github-actions]

Ephemeral environment shutdown and build artifacts deleted.

[sadpandajoe]

🏷️ preset:2022.25