Escape time-like tokens in FTS queries

[DevNexsler]

Changes

• escape unescaped : characters before LanceDB/Tantivy FTS search so natural-language queries like 10:00 and 11:30 stop being parsed as field lookups
• add regression coverage for time-like query tokens against keyword_search

Test Report

• Command: PYTHONPATH=. pytest tests/test_config.py tests/test_prefect_server.py
• Result: ✅ PASS
• Output:

============================= test session starts ==============================
platform linux -- Python 3.12.3, pytest-9.0.2, pluggy-1.6.0
rootdir: /home/danpark/projects/RAG-in-a-Box
configfile: pyproject.toml
plugins: asyncio-1.3.0, anyio-4.13.0
asyncio: mode=Mode.STRICT, debug=False, asyncio_default_fixture_loop_scope=None, asyncio_default_test_loop_scope=function
collected 24 items

tests/test_config.py ............                                        [ 50%]
tests/test_prefect_server.py ............                                [100%]

============================== 24 passed in 1.27s ==============================

• Extra verification: PYTHONPATH=. pytest tests/test_search.py -q → 35 passed

Container Health

• doc-organizer: running, restarts=0, oom=false
• fixed code issue behind repeated Keyword/FTS search failed (degraded to vector-only): Field does not exist: '1'/'10' warnings from time-like query tokens
• separate operational issue remains: OpenRouter embedding requests returned 402 Payment Required in 24h logs