feat: add market sentiment filters (Fear & Greed, CryptoPanic, CryptoQuant)

- SentimentProvider: fetches Fear & Greed Index (free, no key),
  CryptoPanic news sentiment (free key), CryptoQuant exchange
  netflow (free key)
- SentimentData: aggregated should_buy/should_block logic
  - Fear < 30 = buy opportunity, Greed > 80 = block buying
  - Negative news < -0.5 = block buying
  - Exchange outflow = bullish, inflow = bearish
- Integrated into Asian Session RSI strategy as entry filter
- All providers optional — disabled when API key missing
- 14 sentiment tests + 386 total tests passing

3 files changed, 65 insertions, 41 deletions

diff --git a/services/strategy-engine/tests/test_bollinger_strategy.py b/services/strategy-engine/tests/test_bollinger_strategy.py
index 473d9b4..7761f2d 100644
--- a/services/strategy-engine/tests/test_bollinger_strategy.py
+++ b/services/strategy-engine/tests/test_bollinger_strategy.py
@@ -107,12 +107,14 @@ def test_bollinger_squeeze_detection():
     """Tight bandwidth → no signal during squeeze."""
     # Use a strategy with a high squeeze threshold so constant prices trigger squeeze
     s = BollingerStrategy()
-    s.configure({
-        "period": 5,
-        "num_std": 2.0,
-        "min_bandwidth": 0.0,
-        "squeeze_threshold": 0.5,  # Very high threshold to ensure squeeze triggers
-    })
+    s.configure(
+        {
+            "period": 5,
+            "num_std": 2.0,
+            "min_bandwidth": 0.0,
+            "squeeze_threshold": 0.5,  # Very high threshold to ensure squeeze triggers
+        }
+    )
 
     # Feed identical prices → bandwidth = 0 (below any threshold)
     for _ in range(6):
@@ -126,12 +128,14 @@ def test_bollinger_squeeze_detection():
 def test_bollinger_squeeze_breakout_buy():
     """Squeeze ends with price above SMA → BUY signal."""
     s = BollingerStrategy()
-    s.configure({
-        "period": 5,
-        "num_std": 1.0,
-        "min_bandwidth": 0.0,
-        "squeeze_threshold": 0.01,
-    })
+    s.configure(
+        {
+            "period": 5,
+            "num_std": 1.0,
+            "min_bandwidth": 0.0,
+            "squeeze_threshold": 0.01,
+        }
+    )
 
     # Feed identical prices to create a squeeze (bandwidth = 0)
     for _ in range(6):
@@ -149,12 +153,14 @@ def test_bollinger_squeeze_breakout_buy():
 def test_bollinger_pct_b_conviction():
     """Signals near band extremes have higher conviction via %B."""
     s = BollingerStrategy()
-    s.configure({
-        "period": 5,
-        "num_std": 1.0,
-        "min_bandwidth": 0.0,
-        "squeeze_threshold": 0.0,  # Disable squeeze for this test
-    })
+    s.configure(
+        {
+            "period": 5,
+            "num_std": 1.0,
+            "min_bandwidth": 0.0,
+            "squeeze_threshold": 0.0,  # Disable squeeze for this test
+        }
+    )
 
     # Build up with stable prices
     for _ in range(5):
diff --git a/services/strategy-engine/tests/test_ema_crossover_strategy.py b/services/strategy-engine/tests/test_ema_crossover_strategy.py
index 9e48478..67a20bf 100644
--- a/services/strategy-engine/tests/test_ema_crossover_strategy.py
+++ b/services/strategy-engine/tests/test_ema_crossover_strategy.py
@@ -21,9 +21,18 @@ def make_candle(close: float) -> Candle:
     )
 
 
-def _make_strategy(short: int = 3, long: int = 6, pullback_enabled: bool = False) -> EmaCrossoverStrategy:
+def _make_strategy(
+    short: int = 3, long: int = 6, pullback_enabled: bool = False
+) -> EmaCrossoverStrategy:
     s = EmaCrossoverStrategy()
-    s.configure({"short_period": short, "long_period": long, "quantity": "0.01", "pullback_enabled": pullback_enabled})
+    s.configure(
+        {
+            "short_period": short,
+            "long_period": long,
+            "quantity": "0.01",
+            "pullback_enabled": pullback_enabled,
+        }
+    )
     return s
 
 
@@ -103,13 +112,15 @@ def test_ema_reset_clears_state():
 def test_ema_pullback_entry():
     """Crossover detected, then pullback to short EMA triggers signal."""
     strategy = EmaCrossoverStrategy()
-    strategy.configure({
-        "short_period": 3,
-        "long_period": 6,
-        "quantity": "0.01",
-        "pullback_enabled": True,
-        "pullback_tolerance": 0.05,  # 5% tolerance for test simplicity
-    })
+    strategy.configure(
+        {
+            "short_period": 3,
+            "long_period": 6,
+            "quantity": "0.01",
+            "pullback_enabled": True,
+            "pullback_tolerance": 0.05,  # 5% tolerance for test simplicity
+        }
+    )
 
     # Declining prices so short EMA stays below long EMA
     declining = [100, 98, 96, 94, 92, 90, 88, 86, 84, 82]
@@ -129,6 +140,7 @@ def test_ema_pullback_entry():
     # The short EMA will be tracking recent prices; feed a price that pulls back
     # toward it. We use a moderate price to get close to short EMA.
     import pandas as pd
+
     series = pd.Series(list(strategy._closes))
     short_ema_val = series.ewm(span=3, adjust=False).mean().iloc[-1]
     # Feed a candle at approximately the short EMA value
@@ -141,13 +153,15 @@ def test_ema_pullback_entry():
 def test_ema_pullback_cancelled_on_reversal():
     """Crossover detected, then reversal cancels the pending signal."""
     strategy = EmaCrossoverStrategy()
-    strategy.configure({
-        "short_period": 3,
-        "long_period": 6,
-        "quantity": "0.01",
-        "pullback_enabled": True,
-        "pullback_tolerance": 0.001,  # Very tight tolerance — won't trigger easily
-    })
+    strategy.configure(
+        {
+            "short_period": 3,
+            "long_period": 6,
+            "quantity": "0.01",
+            "pullback_enabled": True,
+            "pullback_tolerance": 0.001,  # Very tight tolerance — won't trigger easily
+        }
+    )
 
     # Declining prices
     declining = [100, 98, 96, 94, 92, 90, 88, 86, 84, 82]
@@ -172,12 +186,14 @@ def test_ema_pullback_cancelled_on_reversal():
 def test_ema_immediate_mode():
     """With pullback_enabled=False, original immediate entry works."""
     strategy = EmaCrossoverStrategy()
-    strategy.configure({
-        "short_period": 3,
-        "long_period": 6,
-        "quantity": "0.01",
-        "pullback_enabled": False,
-    })
+    strategy.configure(
+        {
+            "short_period": 3,
+            "long_period": 6,
+            "quantity": "0.01",
+            "pullback_enabled": False,
+        }
+    )
 
     # Declining prices so short EMA stays below long EMA
     declining = [100, 98, 96, 94, 92, 90, 88, 86, 84, 82]
diff --git a/services/strategy-engine/tests/test_macd_strategy.py b/services/strategy-engine/tests/test_macd_strategy.py
index cd24ee0..17dd2cf 100644
--- a/services/strategy-engine/tests/test_macd_strategy.py
+++ b/services/strategy-engine/tests/test_macd_strategy.py
@@ -98,7 +98,9 @@ def test_macd_signal_line_crossover():
     assert len(buy_signals) > 0, "Expected at least one BUY signal"
     # Check that at least one is a signal-line crossover or histogram crossover
     all_reasons = [sig.reason for sig in buy_signals]
-    assert any("crossover" in r for r in all_reasons), f"Expected crossover signal, got: {all_reasons}"
+    assert any("crossover" in r for r in all_reasons), (
+        f"Expected crossover signal, got: {all_reasons}"
+    )
 
 
 def test_macd_conviction_varies_with_distance():