Quantized model updates, switch to recommending TheBloke

[pseudotensor]

• Switch to recommending TheBloke for llama.cpp
• Add prompt types to cover other WizardLM Bloke models.
• Auto-gen kwargs for llama and gpt4all models, should fix model listening to expert settings.

[pseudotensor]

============================================================================================================ short test summary info ============================================================================================================
FAILED tests/test_cli.py::test_cli_langchain_llamacpp - assert 'The cat is sitting on a window sill.' in '\nIt is sitting on a window and looking out at the view of the city outside.\n\nSources [Score | Link]:<p><ul><li>0.39 | <a href="file/user_path_test/./pexels-evg-kowalievska-1170986_sma...
FAILED tests/test_client_calls.py::test_client_chat_nostream_llama7b - AssertionError: assert 'What do you do?' in 'I’m a software engineer with a passion for building scalable and reliable systems. I’ve been working in the tech industry for over 10 years and have experience in a variety of technologies in...
FAILED tests/test_eval.py::test_eval1[True-32-h2oai/h2ogpt-oig-oasst1-512-6_9b] - concurrent.futures.process.BrokenProcessPool: A process in the process pool was terminated abruptly while the future was running or pending.
FAILED tests/test_manual_test.py::test_chat_context - NotImplementedError: MANUAL TEST FOR NOW
FAILED tests/test_manual_test.py::test_upload_one_file - NotImplementedError: MANUAL TEST FOR NOW -- do and ask query of file
FAILED tests/test_manual_test.py::test_upload_multiple_file - NotImplementedError: MANUAL TEST FOR NOW -- do and ask query of files
FAILED tests/test_manual_test.py::test_upload_url - NotImplementedError: MANUAL TEST FOR NOW -- put in URL box https://github.com/h2oai/h2ogpt/ (and ask what is h2ogpt?). Ensure can go to source links
FAILED tests/test_manual_test.py::test_upload_arxiv - NotImplementedError: MANUAL TEST FOR NOW -- paste in arxiv:1706.03762 and ask who wrote attention paper. Ensure can go to source links
FAILED tests/test_manual_test.py::test_upload_pasted_text - NotImplementedError: MANUAL TEST FOR NOW -- do and see test code for what to try
FAILED tests/test_manual_test.py::test_no_db_dirs - NotImplementedError: MANUAL TEST FOR NOW -- Remove db_dirs, ensure can still start up and use in MyData Mode.
FAILED tests/test_manual_test.py::test_upload_unsupported_file - NotImplementedError: MANUAL TEST FOR NOW -- e.g. json, ensure error correct and reasonable, no cascades
FAILED tests/test_manual_test.py::test_upload_to_UserData_and_MyData - NotImplementedError: MANUAL TEST FOR NOW Upload to each when enabled, ensure no failures
FAILED tests/test_manual_test.py::test_chat_control - NotImplementedError: MANUAL TEST FOR NOW save chat, select chats, clear chat, export, import, etc.
FAILED tests/test_manual_test.py::test_subset_only - NotImplementedError: MANUAL TEST FOR NOW UserData, Select Only for subset, then put in whisper.  Ensure get back only chunks of data with url links to data sources.
=========================================================================== 14 failed, 58 passed, 25 skipped, 1 xfailed, 1 xpassed, 2 warnings in 1164.68s (0:19:24) ============================================================================
(h2ollm) jon@pseudotensor:~/h2o-llm$ 

and fixed above tests, broken pool was just OOM doing CPU inference in 32-bit. After removing some things in memory, all good.