Moved numba from common requirements to cuda/rocm specific requirements (#14199)

Signed-off-by: Nishidha Panpaliya <nishidha.panpaliya@partner.ibm.com>

requirements-common.txt

 psutil
 sentencepiece  # Required for LLaMA tokenizer.
 numpy < 2.0.0
-numba == 0.60.0 # v0.61 doesn't support Python 3.9. Required for N-gram speculative decoding.
 requests >= 2.26.0
 tqdm
 blake3

requirements-cuda.txt

 # Common dependencies
 -r requirements-common.txt
 
+numba == 0.60.0 # v0.61 doesn't support Python 3.9. Required for N-gram speculative decoding
+
 # Dependencies for NVIDIA GPUs
 ray[cgraph] >= 2.43.0 # Ray Compiled Graph, required for pipeline parallelism in V1.
 torch == 2.5.1

requirements-rocm.txt

 # Common dependencies
 -r requirements-common.txt
 
+numba == 0.60.0 # v0.61 doesn't support Python 3.9. Required for N-gram speculative decoding
+
 # Dependencies for AMD GPUs
 awscli
 boto3
@@ -11,4 +13,4 @@ peft
 pytest-asyncio
 tensorizer>=2.9.0
 runai-model-streamer==0.11.0
-runai-model-streamer-s3==0.11.0
\ No newline at end of file
+runai-model-streamer-s3==0.11.0