[Leo] Tracking issue for context limit behaviors

[mattmcalister]

Tracking the issues for improving issues around various context limit behaviors

Back-end:

• https://github.com/brave/aichat/issues/11
• https://github.com/brave/aichat/issues/128
• https://github.com/brave/aichat/issues/126

Page context:

• Premium Leo refuses to summarize long text and transcripts #34131
• Full page google doc summarization support #36649
• Match context length of prompt submission form with model being used #34701
• Truncate llama2 prompts and handle truncation errors #32346
• AI Chat: Tweaked article length truncation brave-core#19813
• Implement Smart Truncation for extracted article content #29608
• Conversation too long.. message shown with Premium model, Mixtral model #36179
• The page is too long for Leo.. message still shown even after Leo starts to summarize #36175

User input:

• Add character limit to inputbox #30233
• Give Leo information about when text is being truncated so it can better formulate an answer.  #33006
• Premium users should have a larger context limit #33880
• AI Chat Claude model should allow higher conversation entry limit #34322
• https://github.com/brave/aichat/issues/19
• Suggest questions... button spins when clicked in the Response rate limit modal #36138
• tapping on Suggestion questions once rate limit reached doesn't do anything/displays spinner #36144

More information in the doc https://docs.google.com/document/d/1ThBz6-HB_UufCcD3732VLyvaX7Co7_SYm9AnfM7CdtE/edit

[mattmcalister]

As discussed last week, I'm closing this tracking issue as most of it is done and what's remaining will be reevaluated along with other context-related issues after the latest upgrades land.