#21620: Add DeepSeek embeddings provider for memory search

## Summary - Add DeepSeek embeddings provider for memory search (remote embeddings). - Wire provider into config schema, doctor check, and docs. ## Notes - Uses OpenAI-compatible embeddings endpoint (POST /embeddings) with base URL https://api.deepseek.com/v1. - Requires DEEPSEEK_API_KEY or models.providers.deepseek.apiKey. ## Testing - Added unit tests for DeepSeek embedding provider (patterned after voyage tests). <!-- greptile_comment --> <h3>Greptile Summary</h3> Adds DeepSeek as a new embedding provider for memory search. The implementation follows the existing pattern for OpenAI-compatible providers (similar to Voyage), using the standard `/embeddings` endpoint at `https://api.deepseek.com/v1`. Authentication uses `DEEPSEEK_API_KEY` or `models.providers.deepseek.apiKey`. - Wired into config schema, doctor checks, and documentation - Unit tests added covering client configuration, remote overrides, and model normalization - Integration follows established patterns for remote embedding providers - Missing DeepSeek references in a few type definitions and help text (see inline comments) <h3>Confidence Score: 4/5</h3> - Safe to merge with minor documentation and type definition updates - Implementation follows established patterns correctly and includes good test coverage. Score reflects minor missing type definitions in `src/agents/memory-search.ts` and help text in `src/config/schema.help.ts` that need fixing to maintain consistency across the codebase. - `src/agents/memory-search.ts` and `src/config/schema.help.ts` need type/help text updates to include `deepseek` option _{Last reviewed commit: e1de643} <!-- greptile_other_comments_section --> _{(2/5) Greptile learns from your feedback when you react with thumbs up/down!} <!-- /greptile_comment -->