ggml : add ggml_conv_1d_grouped

[Juste-Leo2]

Overview

This PR adds the ggml_conv_1d_grouped operation (sub-graph) to be used in supporting CCA (Compressed Convolutional Attention) for future support of Zyphra's models (ZAYA1).
(This is a first step towards #22776)

CCA uses a specific convolution system that doesn't seem to be currently implemented in llama.cpp.

Figure 1: Architecture of Compressed Convolutional Attention (CCA), extracted from the ZAYA1 technical report (arXiv:2605.05365).

Here, the added operation amounts to supporting Depthwise Conv and Headwise Conv.

To explain how it works, I made a diagram with matplotlib:

Figure 2: Functioning of the Grouped 1D Convolution operation (example with Groups = 2).

For example, when the group is equal to 2, we have an initial splitting of the tensors with separate convolutions. Then comes a concatenation to get the final tensor back.

Additional information

The code contains a small trick: the idea is to use ggml_view_3d to avoid making lots of memory copies. I took care to keep the code modifications to a minimum to make the review easier.

Here are the results obtained for the tests:

:~/github/llama.cpp/build$ ./test-conv-1d-grouped
Testing ggml_conv_1d_grouped

  TEST: groups=1 (standard conv1d) (IC=128 OC=256 K=3 L=32 G=1 s=1 p=0)
    PASS
  TEST: ZAYA1-8B exact params (IC=1280 OC=1280 K=2 L=16 G=10 s=1 p=0)
    PASS
  TEST: small 2 groups (IC=4 OC=4 K=2 L=8 G=2 s=1 p=0)
    PASS
  TEST: with padding (IC=8 OC=8 K=2 L=16 G=4 s=1 p=1)
    PASS
  TEST: IC != OC (IC=12 OC=6 K=3 L=10 G=3 s=1 p=0)
    PASS
  TEST: stride=2 (IC=8 OC=8 K=2 L=16 G=4 s=2 p=0)
    PASS
  TEST: longer sequence (IC=1280 OC=1280 K=2 L=128 G=10 s=1 p=0)
    PASS

Result: 7 passed, 0 failed

Note: I will do my best to answer questions regarding the implementation. Currently, it uses existing llama.cpp operations, which I think is best for maintainability at the start.

Requirements

• I have read and agree with the contributing guidelines

• AI usage disclosure: YES
  • I used antigravity to understand the operation and make the diagram too.
  • AI was also used to write the code (I am not yet comfortable enough with the ggml C++ syntax) and to make the best possible optimizations.
  • To respect the coding style compared to other implementations.
  • To translate the PR (written in French by my own hands) and improve readability :)