Update command flag handling

`lmcli chat` now supports common prompt flags (model, length, system
prompt, etc)

pkg/cmd/util/util.go

@@ -13,9 +13,9 @@ import (
 	"github.com/charmbracelet/lipgloss"
 )
 
-// fetchAndShowCompletion prompts the LLM with the given messages and streams
-// the response to stdout. Returns all model reply messages.
-func FetchAndShowCompletion(ctx *lmcli.Context, messages []model.Message, callback func(model.Message)) (string, error) {
+// Prompt prompts the configured the configured model and streams the response
+// to stdout. Returns all model reply messages.
+func Prompt(ctx *lmcli.Context, messages []model.Message, callback func(model.Message)) (string, error) {
 	content := make(chan string) // receives the reponse from LLM
 	defer close(content)
 
@@ -109,7 +109,7 @@ func HandleConversationReply(ctx *lmcli.Context, c *model.Conversation, persist
 		}
 	}
 
-	_, err = FetchAndShowCompletion(ctx, allMessages, replyCallback)
+	_, err = Prompt(ctx, allMessages, replyCallback)
 	if err != nil {
 		lmcli.Fatal("Error fetching LLM response: %v\n", err)
 	}