Map Postgres IDENTITY columns safely to Spanner

[SaurabhSingh2115]

Fixes #545474127

Currently, SMT successfully detects legacy PostgreSQL sequences (e.g., SERIAL types) everywhere. However, the dump file parser entirely misses modern SQL-standard IDENTITY configurations (introduced in PostgreSQL 10+). If a user uploads a dump file from a modern database, their auto-increment columns are silently lost.

[codecov]

Codecov Report

❌ Patch coverage is 92.00000% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 51.32%. Comparing base (a046b9a) to head (96f7ae1).

Files with missing lines	Patch %	Lines
sources/postgres/infoschema.go	90.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1396      +/-   ##
==========================================
+ Coverage   51.28%   51.32%   +0.04%     
==========================================
  Files         203      203              
  Lines       22980    23003      +23     
  Branches      552      552              
==========================================
+ Hits        11785    11807      +22     
- Misses      10500    10501       +1     
  Partials      695      695              

Components	Coverage Δ
backend-apis	46.10% <ø> (ø)
backend-library	56.97% <92.00%> (+0.05%)	⬆️
cli	26.62% <ø> (ø)
frontend	39.34% <ø> (+0.02%)	⬆️

Files with missing lines	Coverage Δ
sources/postgres/pgdump.go	73.94% <100.00%> (+0.19%)	⬆️
sources/postgres/infoschema.go	84.86% <90.00%> (+0.16%)	⬆️

... and 1 file with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.