docs(readme): dark theme plots, move examples to end section

- formatter.py: dark background (#1a1a2e) for PID and FFT plots
  with matching text/axis colors
- README: move PID examples to new 'Examples' section at end
- Add terminal output example showing diagnostic format
- Replace example images with dark-themed versions

Author: raylanlin

README.md

@@ -181,16 +181,6 @@ stune pid -i flight.bin -a roll                        # Single axis
 stune pid -i flight.bbl --visual                       # Betaflight
 ```
 
-### PID Step Response Examples
-
-**ArduPilot** (DataFlash `.bin` log):
-
-![ArduPilot PID step response](assets/pid_ardupilot_example.png)
-
-**Betaflight** (Blackbox `.bbl` log):
-
-![Betaflight PID step response](assets/pid_betaflight_example.png)
-
 ### `stune fft`
 
 Frequency-domain vibration analysis with notch filter suggestions.
@@ -371,6 +361,46 @@ Agents call `stune`, get structured tuning recommendations, and can act on them.
 
 ---
 
+## Examples
+
+### Terminal Output
+
+```text
+Platform: Betaflight
+╭──────────────────────────────────────────────────────────────────╮
+│ PID Step Response Analysis                                       │
+╰──────────────────────────────────────────────────────────────────╯
+
+  PITCH: MARGINAL  (steps: 4)
+  ROLL:  MARGINAL  (steps: 1)
+  YAW:   MARGINAL  (steps: 1)
+  Overall: MARGINAL
+
+╭──────────────────────────────────────────────────────────────────╮
+│ FFT Vibration Analysis                                           │
+╰──────────────────────────────────────────────────────────────────╯
+  Vibration: MARGINAL (10.0 m/s²)
+  Freq (Hz)    Amplitude (dB)    Source
+       93.7             -46.5    motor
+    → gyro_notch1_hz: 93.7
+    → gyro_lowpass_hz: 40
+    → acc_lpf_hz: 10
+
+✓ Analysis complete!
+```
+
+### PID Step Response
+
+**ArduPilot** (DataFlash `.bin` log):
+
+![ArduPilot PID step response](assets/pid_ardupilot_example.png)
+
+**Betaflight** (Blackbox `.bbl` log):
+
+![Betaflight PID step response](assets/pid_betaflight_example.png)
+
+---
+
 ## For Humans
 
 Yes, the terminal output is also beautiful. Rich-powered tables, progress bars, color-coded diagnostics — everything you'd expect from a modern CLI. But the architecture underneath is agent-first.

assets/pid_ardupilot_example.png

@@ -490,6 +490,16 @@ def _generate_pid_plot(self, results: Any) -> Optional[str]:
 
         n = len(axes_data)
         fig, axes_plots = plt.subplots(n, 1, figsize=(10, 4 * n), squeeze=False)
+        fig.patch.set_facecolor("#1a1a2e")
+        plt.rcParams.update({
+            "text.color": "#e0e0e0",
+            "axes.labelcolor": "#e0e0e0",
+            "axes.facecolor": "#16213e",
+            "axes.edgecolor": "#4a4a6a",
+            "xtick.color": "#e0e0e0",
+            "ytick.color": "#e0e0e0",
+            "grid.color": "#3a3a5a",
+        })
 
         for i, (axis_name, axis_data) in enumerate(axes_data.items()):
             fft_step = axis_data.get("fft_step", {})
@@ -510,13 +520,13 @@ def _generate_pid_plot(self, results: Any) -> Optional[str]:
             resp = np.array(step_resp)
 
             # Plot step response
-            ax.plot(t, resp, "b-", linewidth=1.5, alpha=0.9, label="Response")
+            ax.plot(t, resp, color="#4fc3f7", linewidth=1.5, alpha=0.9, label="Response")
 
             # Reference lines
-            ax.axhline(0, color="black", linestyle="-", linewidth=0.5, alpha=0.5)
-            ax.axhline(1, color="black", linestyle="--", linewidth=1.0, alpha=0.8, label="Target")
-            ax.axhline(0.1, color="gray", linestyle=":", linewidth=0.8, alpha=0.5)
-            ax.axhline(0.9, color="gray", linestyle=":", linewidth=0.8, alpha=0.5)
+            ax.axhline(0, color="#888888", linestyle="-", linewidth=0.5, alpha=0.5)
+            ax.axhline(1, color="#ff7043", linestyle="--", linewidth=1.0, alpha=0.8, label="Target")
+            ax.axhline(0.1, color="#666666", linestyle=":", linewidth=0.8, alpha=0.5)
+            ax.axhline(0.9, color="#666666", linestyle=":", linewidth=0.8, alpha=0.5)
 
             # Axes limits
             max_time_ms = max(t) if len(t) > 0 else 500
@@ -533,7 +543,7 @@ def _generate_pid_plot(self, results: Any) -> Optional[str]:
 
         plt.tight_layout()
         out_path = self._plot_save("pid_analysis.png")
-        plt.savefig(out_path, dpi=150)
+        plt.savefig(out_path, dpi=150, facecolor=fig.get_facecolor())
         plt.close(fig)
         return out_path
 
@@ -559,6 +569,16 @@ def _generate_fft_plot(
             return None
 
         fig, ax = plt.subplots(figsize=(10, 5))
+        fig.patch.set_facecolor("#1a1a2e")
+        ax.set_facecolor("#16213e")
+        plt.rcParams.update({
+            "text.color": "#e0e0e0",
+            "axes.labelcolor": "#e0e0e0",
+            "axes.edgecolor": "#4a4a6a",
+            "xtick.color": "#e0e0e0",
+            "ytick.color": "#e0e0e0",
+            "grid.color": "#3a3a5a",
+        })
 
         # Full spectrum curve
         if raw_spectrum and raw_spectrum.get("freqs"):
@@ -629,7 +649,7 @@ def _generate_fft_plot(
 
         plt.tight_layout()
         out_path = self._plot_save("fft_spectrum.png")
-        plt.savefig(out_path, dpi=150)
+        plt.savefig(out_path, dpi=150, facecolor=fig.get_facecolor())
         plt.close(fig)
         return out_path