Add bias_riscv

[AtomAlpaca]

添加 bias 的 riscv 优化

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.64%. Comparing base (c9caafd) to head (71f9bc7).
Report is 2 commits behind head on master.

Additional details and impacted files

@@             Coverage Diff             @@
##           master    #6025       +/-   ##
===========================================
+ Coverage   94.69%   95.64%    +0.94%     
===========================================
  Files         769      414      -355     
  Lines      238897   111121   -127776     
===========================================
- Hits       226227   106280   -119947     
+ Misses      12670     4841     -7829     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

The binary size change of libncnn.so (bytes)

architecture	base size	pr size	difference
x86_64	16465128	16465128	0 😘
armhf	7335260	7335260	0 😘
aarch64	10704800	10704800	0 😘

[Copilot]

Pull Request Overview

This PR adds a RISC-V optimized implementation for applying bias in neural network layers, targeting improved performance with FP16 support. Key changes include:

• Addition of a new source file (bias_riscv_zfh.cpp) implementing vectorized FP16 bias application using RISC-V intrinsics.
• Extension of the Bias_riscv class in the header (bias_riscv.h) to declare the FP16-specific function.
• Update of the Bias_riscv implementation (bias_riscv.cpp) to use the new FP16 vectorized function when appropriate.

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
src/layer/riscv/bias_riscv_zfh.cpp	New FP16 vectorized bias addition implementation using __riscv_zvfh intrinsics with a fallback scalar loop.
src/layer/riscv/bias_riscv.h	Declaration of the new forward_inplace_fp16s function within the Bias_riscv class.
src/layer/riscv/bias_riscv.cpp	Modifications to invoke the FP16 optimized function based on runtime conditions.

[nihui]

Thanks for your contribution !