feat(cli): add standalone ov watch subcommand

[yeyitech]

Summary

• Add ov watch subcommand group with list, add, remove, and set subcommands
• Watch tasks are persisted in ~/.openviking/watches.json, decoupled from add-resource processing
• ov watch add <path> [--to <uri>] [--interval <seconds>] registers a path without triggering re-indexing
• ov watch set <path> --interval <seconds> updates the polling interval without re-processing resources
• ov watch remove <path> removes a watch entry
• ov watch list shows all configured watches with path, target URI, interval, and timestamps

Closes #1662

Test plan

• ov watch list shows empty table when no watches configured
• ov watch add /tmp/mydir --to viking://resources/mydir --interval 60 adds entry to watches.json
• ov watch list shows the new entry
• ov watch set /tmp/mydir --interval 120 updates interval without re-indexing
• ov watch remove /tmp/mydir removes the entry
• Duplicate ov watch add for same path raises clear error

🤖 Generated with Claude Code

[github-actions]

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

⏱️ Estimated effort to review: 3 🔵🔵🔵⚪⚪

🏅 Score: 60

🧪 No relevant tests

🔒 No security concerns identified

✅ No TODO sections

🔀 Multiple PR themes Sub-PR theme: Add standalone ov watch subcommand Relevant files: openviking_cli/cli/commands/watch.py openviking_cli/cli/commands/init.py Sub-PR theme: Add summary_instruction parameter for custom semantic generation Relevant files: openviking/client/local.py openviking/server/routers/resources.py openviking/service/resource_service.py openviking/parse/tree_builder.py openviking/storage/queuefs/semantic_dag.py openviking/storage/queuefs/semantic_msg.py openviking/storage/queuefs/semantic_processor.py openviking/utils/resource_processor.py openviking/prompts/templates/semantic/code_summary.yaml openviking/prompts/templates/semantic/document_summary.yaml openviking/prompts/templates/semantic/file_summary.yaml openviking/prompts/templates/semantic/overview_generation.yaml openviking_cli/cli/commands/resources.py openviking_cli/client/base.py openviking_cli/client/http.py Sub-PR theme: Explicitly disable streaming in OpenAI VLM calls Relevant files: openviking/models/vlm/backends/openai_vlm.py Sub-PR theme: Generalize repo URL normalization to all hosts Relevant files: openviking/parse/parsers/code/code.py Sub-PR theme: Improve rerank client handling of empty documents and responses Relevant files: openviking_cli/utils/rerank.py

⚡ Recommended focus areas for review Critical syntax error due to malformed docstring The docstring closing quote is concatenated with result = { on the same line, causing invalid Python syntax. This will prevent the module from importing entirely. """ result = { Incorrect license header New file uses Apache-2.0 instead of required AGPL-3.0 license header per project conventions. # SPDX-License-Identifier: Apache-2.0

[github-actions]

PR Code Suggestions ✨

Explore these optional code suggestions:

Category	Suggestion	Impact
Possible issue	Fix broken docstring in resource processor The docstring closing triple quote was accidentally removed, causing the result dictionary initialization to be interpreted as part of the docstring. This will lead to a NameError when the function tries to use result. Restore the docstring termination and move result = { back to code. openviking/utils/resource_processor.py [88] Process and store a new resource. Workflow: 1. Parse source (writes to temp directory) 2. TreeBuilder moves to AGFS 3. SemanticQueue generates L0/L1 and vectorizes asynchronously - """ result = { + """ + result = { Suggestion importance[1-10]: 9 __ Why: This fixes a critical syntax error where the docstring was not properly closed, causing the result dictionary to be part of the docstring and leading to a NameError.	High