fix(codegen): include plugin-injected fields in condition and orderBy types

[pyramation]

fix(codegen): include plugin-injected fields in condition and orderBy types

Summary

The codegen's buildTableConditionProperties and buildOrderByValues only iterated over table.fields (database columns), so any fields injected by Graphile plugins (e.g., embeddingNearby from VectorSearchPlugin, or EMBEDDING_DISTANCE_ASC/DESC orderBy values) were silently dropped from generated TypeScript types.

This PR fixes three things:

1. scalars.ts — Adds Vector: 'number[]' to SCALAR_TS_MAP so pgvector's Vector scalar resolves to a TypeScript type.
2. input-types-generator.ts — buildTableConditionProperties and buildOrderByValues now accept an optional TypeRegistry and merge any extra fields/values present in the schema's condition/orderBy types that aren't derived from the table's own columns. A new helper collectConditionExtraInputTypes ensures referenced types (e.g., VectorNearbyInput) also get generated.
3. Transitive type resolution — generateCustomInputTypes previously only followed nested types ending with "Input", so enum types like VectorMetric (referenced by VectorNearbyInput.metric) were silently dropped. Now follows all non-scalar types present in the typeRegistry.
4. Tests — 6 new test cases covering plugin-injected condition fields, transitive enum resolution, field deduplication, plugin-injected orderBy values, and backwards compatibility when no TypeRegistry is provided.

All changes are backwards-compatible — the typeRegistry parameter is optional throughout.

Updates since last revision

• Fixed transitive type resolution bug: generateCustomInputTypes (line ~1619) broadened from baseType.endsWith('Input') to !SCALAR_NAMES.has(baseType) && typeRegistry.has(baseType), so referenced enums and other non-scalar types are correctly followed and generated.
• Added dedicated unit test for transitive enum resolution (VectorNearbyInput → VectorMetric).
• Extended existing test to also verify VectorMetric enum values (L2, INNER_PRODUCT, COSINE) appear in generated output.

Review & Testing Checklist for Human

• Broadened transitive resolution safety: The change in generateCustomInputTypes now follows any non-scalar type that exists in the registry (not just *Input). The generatedTypes set prevents double-emission within this function, but types already emitted by earlier codegen stages (e.g., generateEnumTypes for table-column enums) could theoretically be duplicated in the output. Verify this doesn't cause issues with real schemas.
• Type representation for plugin-injected condition fields: Plugin-injected fields use typeRefToTs(field.type) while table-derived fields use scalarToTsType(fieldType) + ' | null'. Verify the nullability semantics are correct for both paths (table fields always get | null, plugin fields get nullability from the TypeRef wrapper).
• End-to-end validation: Run codegen against a real schema that uses graphile-pgvector-plugin and confirm the generated input-types.ts now includes embeddingNearby in condition types, VectorMetric enum, and distance-based orderBy values. The unit tests use mocked registries that may not perfectly match actual PostGraphile output.
• Suggested test plan: Deploy a local PostGraphile instance with pgvector + VectorSearchPlugin, export the schema, run pnpm codegen against it, and inspect the generated input-types.ts for completeness. Alternatively, a follow-up PR (test(codegen): add vector schema integration test for plugin-injected fields #790) adds an integration test with a hand-crafted vector schema fixture.

Notes

• Fixes constructive-io/constructive-planning#663
• Link to Devin Session: https://app.devin.ai/sessions/c3e55218c0a94d99a5d678a374506930
• Requested by: @pyramation

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring