Make unquoted schema lookup case-insensitive for PostgresSQL

[menghaoranss]

Changes proposed in this pull request:

• Make unquoted schema lookup case-insensitive for PostgresSQL

PR Title
Refine PostgreSQL/openGauss schema identifier rules: case-insensitive unquoted lookup with exact quoted match

PR Description

Summary

This PR adjusts schema identifier matching for PostgreSQL and openGauss to support mixed-case stored schema names while preserving quoted-identifier semantics.

Changes

• Updated PostgreSQL identifier case rule provider:
  • Override IdentifierScope.SCHEMA with newInsensitiveRuleSet().getRule(IdentifierScope.SCHEMA).
  • Keep other scopes on the existing lower-case rule set.
• Updated openGauss identifier case rule provider with the same schema-scope override.
• Updated related unit tests to reflect the new schema behavior.

Behavior After Change

For schema identifiers in PostgreSQL/openGauss:

• Unquoted lookup is case-insensitive (LookupMode.NORMALIZED).
• Quoted lookup remains exact (LookupMode.EXACT behavior for quoted identifiers).

Other identifier scopes (table/column/view/etc.) keep their previous behavior.

Why

Previously, PostgreSQL/openGauss schema rules filtered unquoted candidates by lower-case stored names.
That could fail when schema metadata was stored in upper-case or mixed-case forms.
This change ensures unquoted schema lookup can still resolve those names, while quoted lookup remains strict.

Verification

Executed:

• ./mvnw -pl database/connector/dialect/postgresql,database/connector/dialect/opengauss -am -DskipITs -Dspotless.skip=true -Dtest=PostgreSQLIdentifierCaseRuleProviderTest,OpenGaussIdentifierCaseRuleProviderTest test -Dsurefire.failIfNoSpecifiedTests=false
• ./mvnw -pl infra/common -DskipITs -Dspotless.skip=true -Dtest=DatabaseIdentifierContextFactoryTest test -Dsurefire.failIfNoSpecifiedTests=false

All passed.

---

Before committing this PR, I'm sure that I have checked the following options:

• My code follows the code of conduct of this project.
• I have self-reviewed the commit code.
• I have (or in comment I request) added corresponding labels for the pull request.
• I have passed maven check locally : ./mvnw clean install -B -T1C -Dmaven.javadoc.skip -Dmaven.jacoco.skip -e.
• I have made corresponding changes to the documentation.
• I have added corresponding unit tests for my changes.
• I have updated the Release Notes of the current development version. For more details, see Update Release Note

[menghaoranss]

Core conclusion: this PR is safe to merge.

1) Decision

• Merge Verdict: Mergeable
• Reviewed Scope
  • database/connector/dialect/postgresql/.../PostgreSQLIdentifierCaseRuleProvider.java
  • database/connector/dialect/opengauss/.../OpenGaussIdentifierCaseRuleProvider.java
  • Both provider unit tests
  • Schema-lookup parameterized cases in infra/common/.../DatabaseIdentifierContextFactoryTest.java
• Not Reviewed Scope
  • No full local rebuild/e2e run by me; review is based on latest diff, CI, key runtime path, and test semantics
• Need Expert Review
  • No

2) Root-Cause and Fix Mapping

• Root cause is valid: PostgreSQL/openGauss schema lookup previously depended on lower-case matching and could miss mixed/upper stored schema names for unquoted identifiers.
• Fix maps correctly: IdentifierScope.SCHEMA is overridden to insensitive matching, while other scopes keep lower-case behavior.
• Semantics align with official docs:
  • PostgreSQL: unquoted identifiers are folded to lower case; quoted identifiers are case-sensitive.
    PostgreSQL docs
  • openGauss docs (quote_ident) describe quoting when identifier would be case-folded, consistent with the same model.
    openGauss docs

3) Risk Scan (including adversarial pass)

• Cross-dialect/adjacent-feature regression: change is limited to PostgreSQL/openGauss providers; MySQL/Oracle paths are untouched.
• Config-disabled/flag path: METADATA_IDENTIFIER_CASE_SENSITIVITY override in IdentifierCaseRuleResolver remains unchanged and still has higher priority than provider defaults.
• Partial symptom coverage risk: tests now include schema unquoted/quoted positive/negative checks and mixed stored-case schema cases in context factory coverage.

4) Validation Evidence

• CI for head f7e691d...: core checks are green (CI, CheckStyle, Spotless, License all success). Some E2E jobs are skipped by matrix strategy.
• Scope is tight: 5 files, 1 commit, no unrelated changes observed.
• Code-of-conduct alignment:
  • Build/check expectations: CODE_OF_CONDUCT.md lines 18-20
  • Test behavior expectations: lines 96-99

5) Commands Run (exit code)

• curl -sL https://api.github.com/repos/apache/shardingsphere/pulls/38696 (0)
• curl -sL https://api.github.com/repos/apache/shardingsphere/pulls/38696/files?per_page=100 (0)
• curl -sL https://api.github.com/repos/apache/shardingsphere/pulls/38696/commits?per_page=100 (0)
• curl -sL https://api.github.com/repos/apache/shardingsphere/commits/f7e691dbe0a0c0cf86e3de561a1b09cba1a9bb02/check-runs | jq -r ... (0)

Short summary: this is a minimal, semantics-consistent fix with adequate risk coverage and passing CI evidence; merge is recommended.