feat: XLM-RoBERTa tokenizer.json for Jina v3 + BGE-M3 + German NER

Downloaded from FacebookAI/xlm-roberta-large-finetuned-conll03-{english,german}:
  data/jina-v3-hdr/tokenizer.json      8.7 MB  (XLM-RoBERTa 250K vocab)
  data/bge-m3-hdr/tokenizer.json       8.7 MB  (same tokenizer, BGE-M3 = XLM-RoBERTa)
  data/xlm-roberta-de/tokenizer.json   8.7 MB  (German NER variant, same vocab)

Calibration now runs with REAL BPE tokenization:
  Jina v3:   LOADED (XLM-RoBERTa 250K)
  Reranker:  LOADED (Qwen2 151K, from Qwopus tokenizer on disk)

Results with real BPE: ρ still low (0.07 Jina) because avg pairwise
centroid distance ≠ text similarity. Need full MatVec think cycle
(perturb → think → commit) for meaningful calibration, not just
raw centroid lookup averages. The ONNX ground truth path solves this.

https://claude.ai/code/session_019RzHP8tpJu55ESTxhfUy1A

Author: claude

crates/thinking-engine/data/.gitignore

@@ -0,0 +1,2 @@
+*.onnx
+*.onnx_data