feat(gallery): add BGE small English variants

Add F16, Q8_0 and Q4_K_M builds for English retrieval. Pin the source
revision and verify each SHA256 against HF metadata and download headers.
Document query prefixes and the 512-token context.

Assisted-by: Codex:gpt-6

docs/content/features/embeddings.md

@@ -33,6 +33,14 @@ Example gallery models:
 - `qwen3-embedding-8b` - Qwen3 Embedding 8B model  
 - `qwen3-embedding-0.6b` - Qwen3 Embedding 0.6B model
 
+### BGE-small-en-v1.5
+
+Install `bge-small-en-v1.5-f16` for English retrieval with llama.cpp. The gallery
+also offers `bge-small-en-v1.5-q8-0` and `bge-small-en-v1.5-q4-k-m` variants.
+These builds produce 384-dimensional embeddings with a 512-token context.
+Prefix search queries with `Represent this sentence for searching relevant passages: `,
+including the trailing space. Embed documents without this prefix.
+
 ### Example: Using Qwen3-Embedding-4B from Gallery
 
 ```bash