speculative : refactor and add a simpler example

[ggerganov]

cont #10290

• Refactor the speculative decoding into common/speculative. For now, just basic greedy speculation with a single sequence.
• Add a more simple speculative-simple example that uses the new API.

This could be used as a starting point for llama-server speculative decoding support. See the TODOs in the comments for the next steps.

Sample usage:

./bin/llama-speculative-simple \
    -m  ../models/qwen2.5-32b-coder-instruct/ggml-model-q8_0.gguf \
    -md ../models/qwen2.5-0.5b-coder-instruct/ggml-model-q8_0.gguf \
    -f ../../test.txt -c 0 -ngl 99 -ngld 99 --draft 16 --color \
    --sampling-seq k --top-k 1 -fa