fix(review): cache OCR backend, drop asserts, real pipeline_tier, guard zero-page

Address PR #858 review:

- 🔴 OcrProcessor now resolves its backend once and reuses it. Rebuilding per
  call created a fresh GatewayTokenProvider each time -- discarding its M2M-token
  cache, so every OCR'd document fetched a new token -- and a new Mistral client.
- 🔴 build_ocr_backend uses explicit ValueError (not assert, which is stripped
  under `python -O`) for the gateway M2M triple.
- PIPELINE_TIER in the Qdrant payload now reflects the tier that actually
  produced the doc: the registry stamps result.metadata["pipeline_tier"] and the
  processor reads it (was hardcoded "fast", wrong for OCR/structured).
- Escalation now requires classification.page_count > 0, so a zero-page
  (empty/corrupt) PDF isn't pointlessly sent to OCR; documented that a fast
  FAILURE (encrypted/unopenable) is a hard failure and is not OCR-escalated.
- Documented the OCR page_boundaries separator-attribution choice.
- Downgraded the per-document page-boundary / page-assignment INFO logs to debug.

New tests: zero-page no-escalation, pipeline_tier stamping.

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

nextcloud_mcp_server/vector/processor.py

@@ -589,7 +589,7 @@ async def _index_document(
                 # Diagnostic: Log page boundary information if available
                 if "page_boundaries" in file_metadata:
                     page_boundaries = file_metadata["page_boundaries"]
-                    logger.info(
+                    logger.debug(
                         "Page boundaries for %s: %s pages, text length: %s",
                         file_path,
                         len(page_boundaries),
@@ -652,7 +652,7 @@ async def _index_document(
 
             # Diagnostic: Verify page number assignment
             assigned_count = sum(1 for c in chunks if c.page_number is not None)
-            logger.info(
+            logger.debug(
                 "Assigned page numbers to %s/%s chunks for %s",
                 assigned_count,
                 len(chunks),
@@ -926,7 +926,11 @@ async def _index_document(
                     # Decomposition payload keys (design §10.2), additive.
                     payload_keys.PROCESSOR_VERSION: "monolith-v1",
                     payload_keys.PARSED_AT: indexed_at,
-                    payload_keys.PIPELINE_TIER: "fast",
+                    # Actual tier that produced this doc (registry stamps it on
+                    # the result metadata); non-PDF doc types stay "fast".
+                    payload_keys.PIPELINE_TIER: file_metadata.get(
+                        "pipeline_tier", "fast"
+                    ),
                     payload_keys.EMBEDDING_IDENTITY: _embedding_identity,
                     payload_keys.ACL_HASH: _acl_hash,
                     # File-specific metadata (PDF, etc.)