Run DESCRIBE HISTORY in parallel to improve performance

[mars-lan]

🤔 Why?

Each DESCRIBE HISTORY command takes 1~5 seconds to complete. The time adds up quickly when running against a large number of tables.

🤓 What?

• Batch run DESCRIBE HISTORY commands in a thread pool to improve performance.
• Add create_connect_pool util method for code sharing.
• Bump up LIMIT for DESCRIBE HISTORY from 50 to 100 since we're less contained by time.

🧪 Tested?

Verified against a production instance with ~3000 tables and saw close to 10x speed up (took 7 mins to get all last refresh dates).

☑️ Checks

• My PR contains actual code changes, and I have updated the version number in pyproject.toml.

[github-actions]

☂️ Python Coverage

current status: ✅

Overall Coverage

Lines	Covered	Coverage	Threshold	Status
19301	18093	94%	85%	🟢

New Files

No new covered files...

Modified Files

File	Coverage	Status
metaphor/unity_catalog/config.py	100%	🟢
metaphor/unity_catalog/extractor.py	97%	🟢
metaphor/unity_catalog/profile/extractor.py	94%	🟢
metaphor/unity_catalog/utils.py	89%	🟢
TOTAL	95%	🟢

updated for commit: de9f5ea by action🐍

[codecov]

Codecov Report

Attention: Patch coverage is 90.24390% with 4 lines in your changes missing coverage. Please review.

Project coverage is 93.74%. Comparing base (b517c90) to head (ec534f3).

Files	Patch %	Lines
metaphor/unity_catalog/utils.py	85.71%	4 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #917      +/-   ##
==========================================
+ Coverage   93.57%   93.74%   +0.16%     
==========================================
  Files         216      171      -45     
  Lines       19447    19299     -148     
==========================================
- Hits        18197    18091     -106     
+ Misses       1250     1208      -42     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[alyiwang]

LGTM