Apply suggestions from code review

desc/integrals/bounce_integral.py

@@ -197,7 +197,7 @@ class Bounce2D(Bounce):
     single-valued basis functions defined on a bounded subset of ℝ²
     (recall continuous functions on compact sets attain their maximum).
     Still, it suffices to interpolate θ over one branch cut. We choose the
-    branch cut defined by (α, ζ) ∈ [0, 2π]. Here the bound θ ∈ [0, 4π] holds.
+    branch cut defined by (α, ζ) ∈ [0, 2π].
 
     Likewise, α is multivalued. As the field line is followed, the label
     jumps to α ∉ [0, 2π] after completing some toroidal transit. Therefore,
@@ -208,7 +208,7 @@ class Bounce2D(Bounce):
     Recall a jump discontinuity appears as an infinitely sharp cut without
     Gibbs effects. To recover the single-valued θ(α, ζ) from the function
     approximation over one branch cut, at every ζ = 2π ℓ we can add either
-    0 or 2π or 4π to the next cut of θ.
+    an integer multiple of 2π to the next cut of θ.
 
     Examples
     --------

tests/test_integrals.py

@@ -943,9 +943,6 @@ def _one_bump(x, h):
 def _bumpy(x, x_peak=(-0.5, 0, 0.5), h_peak=(0.5, 0.75, 0.25), sigma=0.125):
     """Make |B| with humps.
 
-    Looks fine by eye when plotted, but might not be that realistic of an example?,
-    since no power series centered near center of Gaussian converges (to Gaussian).
-
     Parameters
     ----------
     x : jnp.ndarray