Disable yet another flaky Ollama test point

It is unclear at this time why this test point is unreliable, but it just started failing in the GitHub CI, possibly following some Ollama update.

We are not explicitly promising this behaviour, and the change was not on our side.

tests/tollamaChat.m

@@ -47,6 +47,11 @@ function doGenerateUsingSystemPrompt(testCase)
         end
 
         function extremeTopK(testCase)
+            %% This should work, and it does on some computers. On others, Ollama
+            %% receives the parameter, but either Ollama or llama.cpp fails to
+            %% honor it correctly.
+            testCase.assumeTrue(false,"disabled due to Ollama/llama.cpp not honoring parameter reliably");
+
             % setting top-k to k=1 leaves no random choice,
             % so we expect to get a fixed response.
             chat = ollamaChat("mistral",TopK=1);