feat(deploy): TrueNAS Custom-App + Auto-Deploy, plus aufgelaufene Arbeit

Deployment:
- custom-app.compose.yaml: self-contained Compose fuer TrueNAS "Custom App"
  (absolute Host-Bind-Pfade, postgres:18, pull_policy always, Port 8090)
- scripts/truenas-deploy.sh: Host-Skript create/redeploy via midclt (App
  bleibt unter Apps sichtbar) inkl. Image-Pull + Health-Check
- ci.yml Deploy-Job: laeuft auf ubuntu-latest-Runner, kopiert Deploy-Dateien
  per SSH auf den NAS-Host und triggert truenas-deploy.sh (statt runs-on goldeye)
- compose.yaml/.env.example: postgres:18 (Locale-Match zur Quell-DB), Port 8090
- .gitignore: .agents/, tools/rag/, deploy/truenas/.env (Secrets/Scratch)

Aufgelaufene Feature-Arbeit (verified/Freeze, Migrationen, Import-Triage):
- GerbilOverride/VerifiedGerbil-Endpoints + GerbilSnapshotService + Tests
- EF-Migrationen (ShowInChronicle, Stillborn, BirthOrder, ManualFlag, DSGVO)
- Frontend VerifizierteTierePage + verified-API + e2e-Spec
- diverse Import-/Triage-Skripte und -Tests

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

tools/import/rpro3_mxp.py

@@ -0,0 +1,139 @@
+# -*- coding: utf-8 -*-
+"""Parser für RennmausPro-III `.mxp`-Exporte (aus dem Backup
+`C:\\Users\\gulum\\dev\\RennmausPro IIIb 06.06.26.backup`).
+
+Eine `.mxp` ist ein latin-1-XML-artiges Format mit drei Abschnitten:
+  <stamm>   das Tier selbst (name, clan=Herkunft, sex, geb=Julian Day Number,
+            color="Farbschlag$$Gencode$$...", herk, gew)
+  <baum>    Ahnentafel: <bp>-Zeilen, Felder per <#> getrennt:
+            [_, pos(v/m/vv/vm/...), name, clan/adresse, geb, _, _, color, ...]
+  <litter>  eigene Würfe: <w>-Blöcke mit <date>, <coparent>, <child>-Zeilen
+
+Nutzung:
+  python -X utf8 rpro3_mxp.py <backup.zip-oder-mxp-ordner> [Name ...]
+  # ohne Namen: alle Tiere als kompakte Übersicht
+"""
+import sys, os, zipfile, json, datetime
+
+BACKUP = r"C:\Users\gulum\dev\RennmausPro IIIb 06.06.26.backup"
+
+
+def jdn_to_iso(jdn):
+    try:
+        n = int(str(jdn).strip())
+    except Exception:
+        return None
+    if n <= 0:
+        return None
+    try:
+        return datetime.date.fromordinal(n - 1721425).isoformat()
+    except Exception:
+        return None
+
+
+def _split_color(raw):
+    # "Silberagouti-Schecke$$Aa C- D- E- gg Pp Spsp$$$$$$$$schwarz"
+    parts = (raw or "").split("$$")
+    farbschlag = parts[0].strip() if parts else ""
+    gencode = parts[1].strip() if len(parts) > 1 else ""
+    return farbschlag or None, gencode or None
+
+
+def _tagval(block, tag):
+    key = "<" + tag + ">"
+    for line in block.splitlines():
+        s = line.strip()
+        if s.startswith(key):
+            return s[len(key):].strip()
+    return None
+
+
+def parse_mxp(text):
+    # Split into sections by top-level tags.
+    stamm = {}
+    if "<stamm>" in text:
+        seg = text.split("<stamm>", 1)[1].split("</stamm>")[0] if "</stamm>" in text else text.split("<stamm>", 1)[1]
+        seg = seg.split("<baum>")[0].split("<litter>")[0]
+        name = _tagval(seg, "name")
+        clan = _tagval(seg, "clan")
+        sex = _tagval(seg, "sex")
+        geb = _tagval(seg, "geb")
+        color = _tagval(seg, "color")
+        fs, gc = _split_color(color)
+        stamm = {"name": name, "clan": clan, "sex": sex,
+                 "dob": jdn_to_iso(geb), "farbschlag": fs, "gencode": gc,
+                 "zbnr": _tagval(seg, "zbnr"), "gew": _tagval(seg, "gew")}
+    # Ahnentafel: direct parents (pos == 'v' / 'm')
+    parents = {}
+    for line in text.splitlines():
+        s = line.strip()
+        if not s.startswith("<bp>"):
+            continue
+        f = s.split("<#>")
+        if len(f) < 3:
+            continue
+        pos = f[1].strip()
+        if pos not in ("v", "m"):
+            continue
+        nm = f[2].strip()
+        clan = f[3].strip() if len(f) > 3 else ""
+        geb = f[4].strip() if len(f) > 4 else ""
+        color = next((x for x in f if "$$" in x), "")
+        fs, gc = _split_color(color)
+        parents["Vater" if pos == "v" else "Mutter"] = {
+            "name": nm, "clan": clan.split("##")[0] if clan else clan,
+            "dob": jdn_to_iso(geb), "farbschlag": fs, "gencode": gc}
+    # Würfe
+    litters = []
+    if "<litter>" in text:
+        lit = text.split("<litter>", 1)[1]
+        for wblock in lit.split("<w>")[1:]:
+            date = jdn_to_iso(_tagval(wblock, "date"))
+            co = _tagval(wblock, "coparent") or ""
+            coname = co.split("$$")[0].strip() if co else ""
+            children = []
+            for line in wblock.splitlines():
+                s = line.strip()
+                if s.startswith("<child>"):
+                    cf = s.split("<#>")
+                    csex = cf[2].strip() if len(cf) > 2 else ""
+                    ccolor = next((x for x in cf if "$$" in x), "")
+                    cfs, _ = _split_color(ccolor)
+                    # child name is usually empty in export; keep sex+color
+                    children.append({"sex": csex, "farbschlag": cfs})
+            litters.append({"date": date, "coparent": coname, "children": len(children),
+                            "child_details": children})
+    return {"stamm": stamm, "parents": parents, "litters": litters}
+
+
+def load_all(src=BACKUP):
+    out = {}
+    if os.path.isfile(src) and zipfile.is_zipfile(src):
+        z = zipfile.ZipFile(src)
+        for n in z.namelist():
+            if n.lower().endswith(".mxp"):
+                txt = z.read(n).decode("latin-1", "replace")
+                out[os.path.splitext(os.path.basename(n))[0]] = parse_mxp(txt)
+    else:
+        for n in os.listdir(src):
+            if n.lower().endswith(".mxp"):
+                txt = open(os.path.join(src, n), encoding="latin-1").read()
+                out[os.path.splitext(n)[0]] = parse_mxp(txt)
+    return out
+
+
+if __name__ == "__main__":
+    src = sys.argv[1] if len(sys.argv) > 1 else BACKUP
+    want = [a.lower() for a in sys.argv[2:]]
+    data = load_all(src)
+    for key in sorted(data):
+        if want and key.lower() not in want:
+            continue
+        d = data[key]
+        st = d["stamm"]
+        print(f"\n### {key}  ({st.get('name')}, {st.get('sex')}, *{st.get('dob')})")
+        print(f"    Herkunft: {st.get('clan')} | Farbe: {st.get('farbschlag')} [{st.get('gencode')}]")
+        for role, p in d["parents"].items():
+            print(f"    {role}: {p['name']} ({p.get('clan')}, *{p.get('dob')}) {p.get('farbschlag')} [{p.get('gencode')}]")
+        for l in d["litters"]:
+            print(f"    Wurf {l['date']} x {l['coparent']} -> {l['children']} Junge")