mtmd: fix patch_size initialized to random value in audio models

[ngxson]

Fix #17125

Tests are OK:

[audio]  OK:   llama-mtmd-cli ggml-org/ultravox-v0_5-llama-3_2-1b-GGUF:Q8_0
[audio]  OK:   llama-mtmd-cli ggml-org/Qwen2.5-Omni-3B-GGUF:Q4_K_M
[audio]  OK:   llama-mtmd-cli ggml-org/Voxtral-Mini-3B-2507-GGUF:Q4_K_M

[CISC]

Can't you just have default values in clip_hparams though?

[ngxson]

Can't you just have default values in clip_hparams though?

Either way is possible, but I think it's better to avoid communicating patch_size should have a default value. It's technically should not have a default value. And in the case of audio model, it's more like a dummy value.

[CISC]

Can't you just have default values in clip_hparams though?

Either way is possible, but I think it's better to avoid communicating patch_size should have a default value. It's technically should not have a default value. And in the case of audio model, it's more like a dummy value.

What about at least initializing it to 0 so it'll crash if one forgets for a new modality?

[henk717]

Initializing to 0 breaks the current audio model implementation. It was found because my system initialized it as 0 already resulting in audio models failing to launch at random moments.

[CISC]

Initializing to 0 breaks the current audio model implementation. It was found because my system initialized it as 0 already resulting in audio models failing to launch at random moments.

I meant in clip_hparams (while still setting it to 1 for audio).

[ngxson]

What about at least initializing it to 0 so it'll crash if one forgets for a new modality?

yes and even better, I think there should be default value for other hparams too (added in 4ac2c20)

This should prevent forgetting to initialize one of the params in general.

[LostRuins]

also there are multiple GGML_ASSERT scattered throughout the code that can trip if patch size is an unexpected value even if unused. but this approach should work.