added support for tied weights in qwen pipeline parallelism

[FrankLeeeee]

Motivation

Some Qwen variants have tie_word_embeddings = true, however, the current code only caters to case where its value is false. If this value is true, the last rank cannot tie weights successfully since the embed tokens is a PPMissingLayer on the last rank.

I have added logic to:

1. handle embedding weights send/recv for PP
2. handle weight loading for PP

Checklist

• Format your code according to the Code Formatting with Pre-Commit.
• Add unit tests as outlined in the Running Unit Tests.
• Update documentation / docstrings / example tutorials as needed, according to Writing Documentation.
• Provide throughput / latency benchmark results and accuracy evaluation results as needed, according to Benchmark and Profiling and Accuracy Results.
• For reviewers: If you haven't made any contributions to this PR and are only assisting with merging the main branch, please remove yourself as a co-author when merging the PR.
• Please feel free to join our Slack channel at https://slack.sglang.ai to discuss your PR.