✅ Summary

Fixes: #30179

This pull request resolves a rendering issue in polar plots when using set_yscale("log") together with set_rorigin(...). Previously, when setting a custom radial origin in log scale, the inner circular gridline (representing the minimum radial value) would still be rendered at the center of the polar plot, rather than at the specified origin.

This fix ensures correct rendering of the inner border with proper alignment to the configured rorigin, even under logarithmic radial scaling.

🔧 What Was Changed?

• Fixed incorrect rendering logic in projections/polar.py by updating how rorigin is applied when log scale is used.
• Adjusted transform computation to handle non-zero origins in log scale.

📸 Visual Comparison

🔁 Replace the image link above with the actual URL GitHub gives you when you drag-and-drop your image into the PR box.

🧪 Minimal Code to Reproduce

import numpy as np
import matplotlib.pyplot as plt

r = np.logspace(-1, 1, 500)
theta = np.linspace(0, 2*np.pi, 500)

fig, axs = plt.subplots(1, 2, subplot_kw={'projection': 'polar'}, figsize=(12, 6))

axs[0].set_title("Before Fix (log + rorigin)")
axs[0].set_rscale('log')
axs[0].set_rorigin(-0.5)
axs[0].plot(theta, r)

axs[1].set_title("After Fix (log + rorigin)")
axs[1].set_rscale('log')
axs[1].set_rorigin(-0.5)
axs[1].plot(theta, r)

plt.tight_layout()
plt.show()