test(tools): add screenshot, budget and product checks; enforce file headers

- sim_screenshot.py finds the device drawing in the simulator window,
  crops the screen and compares it with approved golden images.
- check_budgets.py reads key=value DIAG/HARNESS lines (the simulator
  console escapes quotes, so JSON was dropped).
- ciq_test_report.py accepts SDK 9.2.0's real summary line
  `PASSED (passed=N, failed=N, errors=N)`; a captured log is a fixture.
- gen_products.py checks all manifest products are watches with
  API 5.0+ and at least 512 KB for apps.
- lint_project.py rule HDR001: every file carries the Author/Summary
  header (JSON keys, PNG text chunks; generated package-lock exempt).
- make_icon.py --all draws one icon per launcher size in use.
- Golden screenshots for fr255s, venusq2, fr70 and fr965.
- 44 tests, warnings as errors, ruff and mypy --strict clean.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

1 files changed, 38 insertions, 27 deletions

diff --git a/tools/check_budgets.py b/tools/check_budgets.py
index 1e26037..8b55a2b 100644
--- a/tools/check_budgets.py
+++ b/tools/check_budgets.py
@@ -1,10 +1,16 @@
 #!/usr/bin/env python3
+# Author:  cirakg & claude ai
+# Summary: Compare the app's debug diagnostics (start-up time, memory) with the budgets in tools/budgets.json.
 """Compare the app's debug diagnostics with the budgets in tools/budgets.json.
 
-Debug builds print lines like
-    DIAG {"event":"first_frame","startup_ms":123,"mem_used":4567,"mem_total":786432}
-(source/util/Diag.mc). app-diag.sh runs the app on each test device, saves the
-console output to build/diag/<device>.log, then calls this script.
+Two kinds of lines are read from build/diag/<device>.log:
+
+    DIAG first_frame startup_ms=1 mem_used=6984 mem_total=781888
+        printed by the debug build (source/util/Diag.mc) after its first frame;
+    HARNESS first_frame wall_ms=2310
+        added by app-diag.sh: wall-clock time from launching the app with
+        monkeydo until that DIAG line appeared, which includes loading the
+        program (the in-app timer can't see that part).
 
 Usage:  check_budgets.py {perf,mem} LOG [LOG ...] [--budgets FILE] [--summary OUT.json]
 Exit:   0 within budget, 1 over budget, 2 missing or unreadable diagnostics.
@@ -14,38 +20,39 @@ from __future__ import annotations
 
 import argparse
 import json
+import re
 import sys
 from pathlib import Path
-from typing import Any
 
 EXIT_OK, EXIT_OVER, EXIT_MISSING = 0, 1, 2
 
+EVENT_RE = re.compile(r"\b(DIAG|HARNESS)\s+(\w+)((?:\s+\w+=-?\d+(?:\.\d+)?)+)")
+PAIR_RE = re.compile(r"(\w+)=(-?\d+(?:\.\d+)?)")
 
-def read_events(text: str) -> list[dict[str, Any]]:
-    """Every DIAG event in a console log, in order. Malformed lines are skipped."""
-    events = []
-    for line in text.splitlines():
-        start = line.find("DIAG {")
-        if start < 0:
-            continue
-        try:
-            event = json.loads(line[start + len("DIAG ") :])
-        except json.JSONDecodeError:
-            continue
-        if isinstance(event, dict):
-            events.append(event)
+
+def read_events(text: str) -> dict[str, dict[str, float]]:
+    """The first occurrence of each event, keyed as "DIAG.first_frame" etc."""
+    events: dict[str, dict[str, float]] = {}
+    for match in EVENT_RE.finditer(text):
+        key = f"{match.group(1)}.{match.group(2)}"
+        if key not in events:
+            events[key] = {name: float(value) for name, value in PAIR_RE.findall(match.group(3))}
     return events
 
 
-def measure(kind: str, events: list[dict[str, Any]]) -> dict[str, float]:
-    """The values a budget kind looks at, from the first first_frame event."""
-    frame = next((e for e in events if e.get("event") == "first_frame"), None)
+def measure(kind: str, events: dict[str, dict[str, float]]) -> dict[str, float]:
+    """The values one budget kind looks at; empty when the app never drew."""
+    frame = events.get("DIAG.first_frame")
     if frame is None:
         return {}
     if kind == "perf":
-        return {"startup_ms": float(frame["startup_ms"])}
-    total = float(frame["mem_total"])
-    return {"mem_used_fraction": float(frame["mem_used"]) / total if total else 1.0}
+        values = {"startup_ms": frame["startup_ms"]}
+        harness = events.get("HARNESS.first_frame")
+        if harness is not None:
+            values["first_frame_wall_ms"] = harness["wall_ms"]
+        return values
+    total = frame["mem_total"]
+    return {"mem_used_fraction": frame["mem_used"] / total if total else 1.0}
 
 
 def main(argv: list[str] | None = None) -> int:
@@ -63,19 +70,23 @@ def main(argv: list[str] | None = None) -> int:
         device = log.stem
         try:
             values = measure(args.kind, read_events(log.read_text(encoding="utf-8", errors="replace")))
-        except (OSError, KeyError, ValueError) as err:
+        except (OSError, KeyError) as err:
             print(f"[{device}] cannot read diagnostics: {err}")
             worst = max(worst, EXIT_MISSING)
             continue
         if not values:
-            print(f"[{device}] no DIAG first_frame line (did the app start? is this a debug build?)")
+            print(f"[{device}] no 'DIAG first_frame' line (did the app start? is this a debug build?)")
             worst = max(worst, EXIT_MISSING)
             continue
         summary[device] = values
         for name, limit in budgets.items():
+            if name not in values:
+                print(f"[{device}] {name}: not measured")
+                worst = max(worst, EXIT_MISSING)
+                continue
             value = values[name]
             over = value > limit
-            print(f"[{device}] {name} = {value:.3g} (budget {limit:.3g}) {'OVER BUDGET' if over else 'ok'}")
+            print(f"[{device}] {name} = {value:.4g} (budget {limit:.4g}) {'OVER BUDGET' if over else 'ok'}")
             if over:
                 worst = max(worst, EXIT_OVER)
     if args.summary: