[DOC] Minor tweaks to CONTRIBUTING.md

[acosmicflamingo]

Report incorrect documentation

There are just very few tweaks that need to be made to CONTRIBUTING.md:

1. CLA hyperlink points to non-existent URL:

-6. Sign the [CLA](https://github.com/NVIDIA/numba-cuda/CLA.md)
+6. Sign the [CLA](https://github.com/NVIDIA/numba-cuda/blob/main/CLA.md)

2. Update pip install commands to work with pyproject.toml file that now utilizes the [dependency-groups] section, a feature first introduced in pip's 25.1 release:

-pip install -e ".[test-cu12]"
+pip install -e . --group test-cu12
 
-pip install -e ".[test-cu13]"
+pip install -e . --group test-cu13

Prior to dependency group usage, test-cu12 and test-cu13 were in the [project.optional-dependencies] section of pyproject.toml file. When I was trying to run those commands, I was puzzled to see pip not finding the targets, and then put together what was going on.

[acosmicflamingo]

The following PR will fix the problems addressed in this ticket.

[acosmicflamingo]

This PR is all that's needed for me to successfully run a test using CUDA 12.x:

~/Development/numba-cuda/testing$ python -m pytest -k test_nvjitlink_jit_with_linkable_code_lto_dump_assembly
========================================================== test session starts ==========================================================
platform linux -- Python 3.13.13, pytest-8.4.2, pluggy-1.6.0
benchmark: 5.2.3 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: /home/agbal/Development/numba-cuda/testing
configfile: pytest.ini
plugins: xdist-3.8.0, benchmark-5.2.3
collected 2060 items / 2058 deselected / 2 selected                                                                                     

cudadrv/test_nvjitlink.py ss 

Also just updated drivers to latest and was able to successfully run test using an environment with CUDA 13.x version:

~/Development/numba-cuda/testing$ python -m pytest -k test_cuda_matmul
========================================================== test session starts ==========================================================
platform linux -- Python 3.13.13, pytest-8.4.2, pluggy-1.6.0
benchmark: 5.2.3 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: /home/agbal/Development/numba-cuda/testing
configfile: pytest.ini
plugins: xdist-3.8.0, benchmark-5.2.3
collected 2060 items / 2059 deselected / 1 selected                                                                                     

cudapy/test_matmul.py .                                                                                                           [100%]

================================================== 1 passed, 2059 deselected in 3.01s ===================================================