fix(sqla): convert prequery results to native python types

[villebro]

SUMMARY

When limiting a query with top n series on a db that doesn't support subqueries (e.g. Druid), numeric values are sometimes returned as numpy types. For instance, integer values are returned as numpy.int64 instead of int. If the groupby column type is set to String type, despite the expression returning a numeric value, this causes the following SQLAlchemy compilation error:

This is due to SQLAlchemy attempting to perform a string replacement operation on the numeric value to escape hyphens during query compilation. However, by ensuring that the value is first converted into its native Python equivalent (np.inp64 to int), SQLAlchemy is able to ignore the incorrect target type (in this case String), and is able to successfully compile the query:

TESTING INSTRUCTIONS

1. Use a database that doesn't support subqueries, e.g. Druid
2. Create a dataset with a calculated column that returns a numeric value but set it to "STRING" type
3. Create a chart with a limit on the series count
4. Notice how the query raises a 500 when fetching data

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

[zhaoyongjie]

LGTM

[codecov]

Codecov Report

Merging #17195 (e2ea5db) into master (4c708af) will decrease coverage by 0.07%.
The diff coverage is 66.66%.

@@            Coverage Diff             @@
##           master   #17195      +/-   ##
==========================================
- Coverage   76.91%   76.84%   -0.08%     
==========================================
  Files        1039     1039              
  Lines       55568    55571       +3     
  Branches     7570     7570              
==========================================
- Hits        42742    42702      -40     
- Misses      12576    12619      +43     
  Partials      250      250              

Flag	Coverage Δ
hive	81.46% <66.66%> (+<0.01%)	⬆️
mysql	81.92% <66.66%> (+<0.01%)	⬆️
postgres	81.93% <66.66%> (+<0.01%)	⬆️
presto	?
python	82.28% <66.66%> (-0.15%)	⬇️
sqlite	81.60% <66.66%> (+<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	86.04% <0.00%> (-1.31%)	⬇️
superset/utils/core.py	90.14% <100.00%> (+0.04%)	⬆️
superset/db_engine_specs/presto.py	84.30% <0.00%> (-6.07%)	⬇️
superset/models/core.py	89.26% <0.00%> (-0.74%)	⬇️

---

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 4c708af...e2ea5db. Read the comment docs.

[sadpandajoe]

🏷️ 2021.42