liechtenstein_py: pass omega_max = Lambda/beta to FiniteTempBasis

Lambda = 2 max(8t, b_eff) beta is the dimensionless cutoff beta*omega_max,
but c9 passed it as wmax, so omega_max was 800 instead of 16 and the basis
had 64 functions instead of 37. J_0(mu = 0) changes from -0.29545914 to
-0.29545899, and J_0(mu) changes by at most 7.3e-7 over the plotted range.

Refs #14

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

Author: shinaoka

src/liechtenstein_py.ipynb

@@ -201,7 +201,7 @@
    "outputs": [],
    "source": [
     "Lambda = 2*max(8*t, b_eff) * beta\n",
-    "basis = sparse_ir.FiniteTempBasis(statistics='F', beta=beta, wmax=Lambda, eps=1e-7)\n",
+    "basis = sparse_ir.FiniteTempBasis(statistics='F', beta=beta, wmax=Lambda/beta, eps=1e-7)\n",
     "l = liechtenstein(basis, t, beta, nk, b_eff)"
    ]
   },