context : allow cache-less context for embeddings

[ggerganov]

target #12799

There is no need to create a KV cache when using embedding models such as BERT. This saves memory compared to master.

API Changes

• The llama_encode() method is now the recommended way to compute embeddings and rerank.
• llama_decode() can still be used to compute embeddings as before.
• For embedding models such as BERT, llama_decode() fallbacks to llama_encode() and prints a warning.

In short, whenever the KV cache is not needed - use llama_encode(). Otherwise - use llama_decode(). The changes are backwards compatible.

[ggerganov]

I'll work on rebasing and merging this next - it should be a good improvement for embedding models by reducing the allocated memory during inference.

[aviallon]

Thanks for your awesome work Georgi.
Is there a donation / sponsoring page btw?