Trim redundant tests from the non-finite dissimilarity fix

Drop the Task timeout helper, since ExUnit already times out a hanging
test. Drop the all-NaN case, which exercises the same path as the NaN
coordinate one, and the jit_apply case, since jit is already covered
elsewhere in the file. Drop the finite data case, which compared fit/2
against itself once the fallback path was removed.

Author: RicardoSantos-99

test/scholar/cluster/hierarchical_test.exs

@@ -288,18 +288,6 @@ defmodule Scholar.Cluster.HierarchicalTest do
     # always mutually nearest to each other, guaranteeing at least one merge every round.
     # When it happens the loop now stops instead of running forever, and the merges it could
     # not make are reported as NaN dissimilarities with clade -1 and size 0.
-    #
-    # Every case is wrapped in a Task with an explicit timeout so a regression fails fast
-    # with a clear message instead of hanging the whole test run.
-    defp fit_within(data, opts, ms \\ 5_000) do
-      task = Task.async(fn -> Hierarchical.fit(data, opts) end)
-
-      case Task.yield(task, ms) || Task.shutdown(task, :brutal_kill) do
-        {:ok, result} -> result
-        nil -> flunk("Hierarchical.fit did not terminate within #{ms}ms")
-      end
-    end
-
     test "an infinite dissimilarity between the last two clades reports an incomplete merge" do
       # Point 3 is infinitely far from everyone. Once points 0, 1 and 2 (mutually close)
       # have merged, only two clades remain: {0, 1, 2} and {3}, at distance infinity, and
@@ -312,7 +300,7 @@ defmodule Scholar.Cluster.HierarchicalTest do
           [:infinity, :infinity, :infinity, 0.0]
         ])
 
-      model = fit_within(d, dissimilarity: :precomputed, linkage: :single)
+      model = Hierarchical.fit(d, dissimilarity: :precomputed, linkage: :single)
 
       assert model.num_points == 4
       # The two finite merges are made and kept, in ascending order.
@@ -329,38 +317,12 @@ defmodule Scholar.Cluster.HierarchicalTest do
       # which is why an incomplete merge is identified by its clade and size, not by NaN.
       x = Nx.tensor([[0.0, 0.0], [0.1, 0.0], [0.2, 0.0], [5.0, 5.0], [5.1, 5.0], [:nan, 0.0]])
 
-      model = fit_within(x, [])
+      model = Hierarchical.fit(x)
 
       assert model.num_points == 6
       assert Nx.to_number(Nx.all(Nx.not_equal(model.clades, -1))) == 1
       assert Nx.to_number(Nx.all(Nx.greater(model.sizes, 0))) == 1
     end
-
-    test "an all-NaN input still makes every merge" do
-      x = Nx.broadcast(Nx.tensor(:nan), {6, 2})
-
-      model = fit_within(x, [])
-
-      assert model.num_points == 6
-      assert Nx.to_number(Nx.all(Nx.not_equal(model.clades, -1))) == 1
-      assert Nx.to_number(Nx.all(Nx.greater(model.sizes, 0))) == 1
-    end
-
-    test "does not change the result on finite data" do
-      # Aborting only ever triggers when a round makes zero progress, which cannot happen
-      # for finite dissimilarities, so this must match plain `fit/2` exactly.
-      data = Nx.tensor([[1, 5], [2, 5], [1, 4], [4, 5], [5, 5], [5, 4], [1, 2], [1, 1], [2, 1]])
-
-      for linkage <- [:average, :complete, :single, :ward, :weighted] do
-        assert fit_within(data, linkage: linkage) == Hierarchical.fit(data, linkage: linkage)
-      end
-    end
-
-    test "works with jit_apply" do
-      x = Nx.tensor([[0.0, 0.0], [1.0, 0.0], [0.0, 1.0], [:nan, :nan], [5.0, 5.0]])
-
-      assert Nx.Defn.jit_apply(&Hierarchical.fit/1, [x]) == Hierarchical.fit(x)
-    end
   end
 
   describe "errors" do