[BUG] Ollama Keep Alive "-1" not working

[sebaxakerhtc]

Describe the bug
A clear and concise description of what the bug is.
Ollama's keep alive setting supports "-1" value, so keep model forever.
Flowise's ChatOllama doesn't support this value and set the default 4 minutes.

To Reproduce
Steps to reproduce the behavior:

1. Add 'ChatOllama' to workflow
2. Click on 'ChatOllama's additionals settings'
3. Set the value "-1" for 'Keep Alive'
4. See "4 minutes from now" with 'ollama ps' command after model loaded

Expected behavior
Expected to see "forever from now" with 'ollama ps' command after model loaded

Screenshots

Flow
Not applicable.

Setup

• Installation [docker`]
• Flowise Version [2.1.5]
• OS: [Windows]
• Browser [any]

Additional context
Add any other context about the problem here.

[sebaxakerhtc]

A half-related discussion, where we can understand why it's important to have a single model forever in VRAM. Just my use-case