v0.0.98

ci: fix build

- windows amd64
- remove darwin ccache

Signed-off-by: thxCode <thxcode0824@gmail.com>

v0.0.97

refactor: images api

Signed-off-by: thxCode <thxcode0824@gmail.com>

v0.0.96

refactor: tool calling

Signed-off-by: thxCode <thxcode0824@gmail.com>

v0.0.95

chore: bump llama.cpp

Signed-off-by: thxCode <thxcode0824@gmail.com>

v0.0.94

refactor: metric prefix

Signed-off-by: thxCode <thxcode0824@gmail.com>

v0.0.93

refactor: metal patch

Signed-off-by: thxCode <thxcode0824@gmail.com>

v0.0.92

fix: embeddings/rerank processing

Signed-off-by: thxCode <thxcode0824@gmail.com>

v0.0.91

ci: adjust cuda build

Signed-off-by: thxCode <thxcode0824@gmail.com>

v0.0.90

fix: failed decoding llava 1.6

Signed-off-by: thxCode <thxcode0824@gmail.com>

v0.0.89

ci: ccache

Signed-off-by: thxCode <thxcode0824@gmail.com>