feat(refactor): Document refactoring progress and phases in markdown

feat(scripts): Add backfill script for content_hash in cache tables

feat(scripts): Create recompute script for analysis_cache population

test(tests): Implement comprehensive tests for analysis module functions

fix(tests): Update CLI tests to assert errors on stderr instead of stdout

fix(tests): Adjust MCP integration tests to pass context parameter correctly

fix(tests): Modify service tests to return hash on save functions for consistency

finn_eiendom/feedback.py

@@ -3,7 +3,11 @@
 import logging
 from typing import Any
 
+from .cache import delete_feedback as cache_delete_feedback
+from .cache import get_feedback as cache_get_feedback
+from .cache import get_feedback_by_verdict
 from .cache import init_db
+from .cache import save_feedback as cache_save_feedback
 from .config import FINN_CACHE_PATH
 
 logger = logging.getLogger(__name__)
@@ -21,15 +25,7 @@ def save_feedback(finnkode: str, verdict: str, notes: str | None = None) -> dict
         Dict with saved feedback details
     """
     conn = init_db(FINN_CACHE_PATH)
-
-    # TODO: implement via feedback table in cache.py
-    # For now, return a success response
-    return {
-        "finnkode": finnkode,
-        "verdict": verdict,
-        "notes": notes,
-        "saved": True,
-    }
+    return cache_save_feedback(conn, finnkode, verdict, notes)
 
 
 def get_feedback(finnkode: str) -> dict[str, Any] | None:
@@ -42,9 +38,21 @@ def get_feedback(finnkode: str) -> dict[str, Any] | None:
         Feedback dict if exists, else None
     """
     conn = init_db(FINN_CACHE_PATH)
+    return cache_get_feedback(conn, finnkode)
 
-    # TODO: implement via feedback table in cache.py
-    return None
+
+def get_feedback_by_verdict_impl(verdict: str, limit: int = 100) -> list[dict[str, Any]]:
+    """Retrieve all stored feedback with a given verdict.
+
+    Args:
+        verdict: Verdict to filter by
+        limit: Max results to return
+
+    Returns:
+        List of feedback dicts
+    """
+    conn = init_db(FINN_CACHE_PATH)
+    return get_feedback_by_verdict(conn, verdict, limit=limit)
 
 
 def delete_feedback(finnkode: str) -> dict[str, Any]:
@@ -57,9 +65,4 @@ def delete_feedback(finnkode: str) -> dict[str, Any]:
         Status dict
     """
     conn = init_db(FINN_CACHE_PATH)
-
-    # TODO: implement via feedback table in cache.py
-    return {
-        "finnkode": finnkode,
-        "deleted": True,
-    }
+    return cache_delete_feedback(conn, finnkode)