RHOAIENG-30247: chore(deps): bump PyTorch packages across all Pipfiles (torch 2.6.0, torchvision 0.21.0, tensorboard 2.17.1) (#1462)

* chore(deps): update PyTorch source URLs to use cu124 across all Pipfiles
* also update pytorch_triton_rocm-3.2.0

This is the lowest Cuda version that torch 2.6.0 wheel is built for.
See https://pytorch.org/get-started/previous-versions/#linux-and-windows-1

* Update Pipfile.lock files by piplock-renewal.yaml action

---------

Co-authored-by: GitHub Actions <github-actions[bot]@users.noreply.github.com>

Author: jiridanek

codeserver/ubi9-python-3.11/Pipfile

@@ -5,7 +5,7 @@ verify_ssl = true
 
 [[source]]
 name = "pytorch"
-url = "https://download.pytorch.org/whl/cu118"
+url = "https://download.pytorch.org/whl/cu124"
 verify_ssl = true
 
 [dev-packages]

codeserver/ubi9-python-3.11/Pipfile.lock

@@ -4,7 +4,7 @@ verify_ssl = true
 name = "pypi"
 
 [[source]]
-url = "https://download.pytorch.org/whl/cu121"
+url = "https://download.pytorch.org/whl/cu124"
 verify_ssl = true
 name = "pytorch"
 
@@ -13,8 +13,8 @@ name = "pytorch"
 [packages]
 # PyTorch packages
 tensorboard = "~=2.17.1"
-torch = {version = "~=2.4.0", index = "pytorch"}
-torchvision = {version = "~=0.19.0", index = "pytorch"}
+torch = {version = "~=2.6.0", index = "pytorch"}
+torchvision = {version = "~=0.21.0", index = "pytorch"}
 
 # Datascience and useful extensions
 boto3 = "~=1.35.14"