Bisect provider-rejected batches in triage and deep assessment

DeepSeek's content filter rejects a whole request (400 "Content Exists
Risk") when one article trips it, which cost the other articles in the
batch their assessment and retried them every run. New curate/batch.rs
runs both stages through a bisecting runner: a rejected batch is split
until the offending article is isolated, that article is retried once on
the editor provider when it is a different one, and a still-rejected
article is recorded as a provider_rejected assessment row so it is not
retried for assessment_reuse_days. Cache reuse accepts rows from either
configured model. Each stage logs reused/requested/rejected counts, the
curation: line shows rejections when non-zero, explain prints them, and
the llm_assess span reports the deep-set size.

Implemented by a Claude agent from an orchestrator brief; verified
fmt/clippy(-W dead_code)/test green (354 lib tests).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01A1rCLQeKBgnBo3oTgHuTMe

src/curate/embedding.rs

@@ -416,8 +416,8 @@ pub fn decode_blob(bytes: &[u8], dimension: usize) -> Result<Vec<f32>, Embedding
         });
     }
     let mut vector = Vec::with_capacity(dimension);
-    for chunk in bytes.chunks_exact(4) {
-        let value = f32::from_le_bytes([chunk[0], chunk[1], chunk[2], chunk[3]]);
+    for chunk in bytes.as_chunks::<4>().0 {
+        let value = f32::from_le_bytes(*chunk);
         if !value.is_finite() {
             return Err(EmbeddingError::NonFinite);
         }