add Q6 support

[lihaoyang-amd]

1. Support Q6
2. Optimize the conversion of pack_add, etc., the conversion address appears redundant, using union is more concise and safe;
   code:

  nv_bfloat162* tA = reinterpret_cast<nv_bfloat162*>(&a);
  nv_bfloat162* tB = reinterpret_cast<nv_bfloat162*>(&b);
  nv_bfloat162 tR = __hmax2(*tA, *tB);
  return *(reinterpret_cast<int*>(&tR));
->
  bf162_int_union A, B, R;
  A.i = a;
  B.i = b;
  R.bf2 = __hmax2(A.bf2, B.bf2);
  return R.i;

3. cast float function overloading; we do not need to consider expanding; overloading to achieve a more concise, no template overhead, compilation will be faster!
   code:

__quickreduce_device_inline__ float T2float_cast(half a) {
  return __half2float(a);
}

__quickreduce_device_inline__ float T2float_cast(nv_bfloat16 a) {
  return __bfloat162float(a);
}

4. Removed some unneeded functions

T2uchar_cast
T2int_cast
group_max_min

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀