Maybe add a status bar?

[gitwittidbit]

I'm hosting Perplexica locally along with Ollama which Perplexity uses for answer generation.

As my private instance is used rather infrequently, most of the time, Ollama needs to load the llm first - which, depending on its size, takes a while. And during that wait, I don't know whether everything is working or whether something is amiss.

Therefore, it would be great, if Perplexity could give some feedback where things are atm, e.g. "contacting model provider", "model provider is loading up model", "searching the internet", "providing found information to llm", etc.

(I'm aware that I can have Perplexity have Ollama keep the model in memory, but my GPU's memory is limited and I don't want to block it with a model that isn't used all the time.)