fix(usage): embed query once across doc_types; address review round 1

Round-1 claude-review findings:

- 🔴 Multi-doc_type search billed N embedding calls as 1. nc_semantic_search
  loops search() once per doc_type on one BM25HybridSearchAlgorithm instance,
  and each call re-embedded the query, so only the last query_token_count was
  recorded. Cache the dense embedding per query on the (per-request) instance
  so the query is embedded — and metered — exactly once regardless of how many
  doc_types are searched. This also removes the redundant per-type embed work
  and avoids billing a user N× for one logical query.
- 🟡 Ollama embed() now delegates to embed_with_usage() so single and batch
  embeds use the same /api/embed endpoint (was the legacy /api/embeddings),
  keeping _detect_dimension and other embed() callers consistent.
- 🟢 round() instead of truncating int() when coercing provider-reported token
  counts (forward-compatible if a provider ever returns a float).

Tests: per-instance query-embedding cache (embedded once across 3 doc_types;
re-embeds on a different query).

Deferred (stated on the PR): mistral/openai single-embed dual path (changes
tested error/request semantics on the cloud-critical path — separate refactor),
bedrock boto3 sync-in-async (pre-existing; no new invoke_model calls per doc).

Deck #67.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

nextcloud_mcp_server/providers/openai.py

@@ -233,7 +233,7 @@ class OpenAIProvider(Provider):
         # gives an int, but test doubles / partial responses can surface a
         # non-numeric attribute — fall back to the estimate there.
         tokens = (
-            int(total_tokens)
+            round(total_tokens)
             if isinstance(total_tokens, (int, float))
             else self._estimate_tokens(batch)
         )