Set hpu-extension to 6b2f6fb (#1684)

Upgrading vllm-hpu-extension with change introducing the fix for
unsupported block_softmax_adjustment in fp16 precision

Author: ksmusz

requirements/hpu.txt

@@ -7,7 +7,7 @@ ray
 triton==3.1.0
 setuptools>=77.0.3
 setuptools-scm>=8
-vllm-hpu-extension @ git+https://github.com/HabanaAI/vllm-hpu-extension.git@cd79204
+vllm-hpu-extension @ git+https://github.com/HabanaAI/vllm-hpu-extension.git@6b2f6fb
 
 # Dependencies for HPU vllm docker image
 datasets