Add noisy issue filters

[Photon101]

Summary

Adds transparent controls for skipping noisy or incentive-misaligned issue results:

• repeated --exclude-repo owner/name
• --exclude-title-regex
• --exclude-bounty-like
• skipped counts in text and JSON output

This keeps scans focused on durable maintainer-facing work when the goal is credibility rather than token/reaction/promotion tasks.

Fixes #5

Verification

python3 -m pytest -q
python3 -m py_compile src/hubsignal/cli.py tests/test_cli.py
hubsignal --help
PYTHONPATH=src python3 -m hubsignal --query "is:issue is:open label:\"good first issue\" language:Python -label:claimed" --limit 5 --min-stars 1 --exclude-archived --exclude-forks --exclude-bounty-like --exclude-repo Scottcjn/rustchain-bounties --format json

Results:

7 passed in 0.05s

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 052b45a5ae

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Match earn as a whole word

Because is_bounty_like checks each pattern with plain substring matching, this earn entry also matches common non-bounty words such as “learn” and “learning”. When users run with --exclude-bounty-like, legitimate good-first/onboarding issues with titles or bodies like “Learn the plugin API” will be skipped as bounty-like, so this should use a word-boundary/phrase match instead of a raw substring.

Useful? React with 👍 / 👎.