Bump the production-dependencies group across 1 directory with 3 updates

[dependabot]

Bumps the production-dependencies group with 3 updates in the / directory: finch, gemini_ex and hammer.

Updates finch from 0.19.0 to 0.20.0

Changelog

Sourced from finch's changelog.

v0.20.0 (2025-07-04)

Enhancements

• Support manual pool termination #299
• Refactor HTTP1 pool state for better maintainability #308
• Add :supported_groups to list of TLS options #307
• Be more explicit about the :default pool in documentation #314
• Upgrade nimble_options to document deprecations #315

Bug Fixes

• Fix Finch.stream_while/5 on halt for both HTTP/1 and HTTP/2 #320
• Return accumulator when Finch.stream/5 and Finch.stream_while/5 fail #295
• Fix documentation reference for get_pool_status/2 #301

Other

• Upgrade CI VM to Ubuntu 24 #321
• CI housekeeping: support Elixir 1.17/Erlang OTP 27, bump Credo and deps #303
• Update GitHub CI badge URL #304

Commits

• b0cde42 Merge pull request #322 from sneako/release-0.20.0
• 24bdf93 release 0.20.0
• dd276fa Merge pull request #320 from wojtekmach/wm-http1-http2-halt
• 7e6c170 Fix Finch.stream_while/5 on halt on http1 + http2
• 56d902f Merge pull request #321 from sneako/fix-ci
• 05607f9 update deps
• c15bde0 upgrade ci vm to ubuntu 24
• 3cf1406 Merge pull request #315 from PragTob/protocol-docs
• 049d284 Merge pull request #314 from PragTob/more-explicit-about-default-pools
• f160024 Upgrade nimble_options to document deprecations
• Additional commits viewable in compare view

Updates gemini_ex from 0.0.2 to 0.8.6

Changelog

Sourced from gemini_ex's changelog.

[0.8.6] - 2025-12-20

Changed

• Default models updated to Gemini 2.5: All defaults now use current-generation models
  • Vertex AI default: gemini-2.0-flash-lite → gemini-2.5-flash-lite
  • Universal default: gemini-2.0-flash-lite → gemini-2.5-flash-lite
• Updated all documentation examples to use gemini-2.5-flash instead of gemini-2.0-flash-exp
• Updated test helpers (universal_model, structured_output_model) to use 2.5 models
• Refreshed model comparison examples in examples/07_model_info.exs

Fixed

• Documentation consistency: All guides, examples, and docstrings now reference current-generation models
• Test model references updated across live session, function calling, and system instruction tests

Notes

• Gemini 2.0 models remain fully supported and available via explicit model selection
• All gemini-2.0-* model keys retained in manifest for backward compatibility
• Context caching still supports both 2.0 and 2.5 model versions

[0.8.5] - 2025-12-18

Added

• response_json_schema structured outputs support with structured_json/2 defaulting to JSON Schema
• Built-in tools serialization for GenerateContent (googleSearch, urlContext, codeExecution)
• Gemini 3 thinking levels :minimal/:medium and :ultra_high media resolution
• Veo 3.x video inputs (image, last_frame, reference_images, video extension, resolution) and Gemini API video generation
• New model registry entries (Gemini 3 Flash preview, 2.0/2.5 variants, flash image, native audio previews, deep research)

Changed

• person_generation defaults to :allow_none (with :dont_allow alias) for image/video config
• total_reasoning_tokens renamed to total_thought_tokens in Interactions usage
• Removed object field from Interaction

Fixed

• Vertex Interactions get/cancel/delete paths now include project/location
• Documentation/examples refreshed for Gemini 3 thinking levels, JSON Schema outputs, and built-in tools

[0.8.4] - 2025-12-13

Added

• Comprehensive examples suite: 10 numbered examples covering all major features
  • 01_basic_generation.exs - Simple/configured/creative vs precise generation
  • 02_streaming.exs - Real-time streaming with timing analysis
  • 03_chat_session.exs - Multi-turn conversations with context retention
  • 04_embeddings.exs - Single/batch embeddings, similarity, task types
  • 05_function_calling.exs - Tool registration and automatic execution
  • 06_structured_outputs.exs - JSON schema, entity extraction, classification
  • 07_model_info.exs - Model listing, details, comparison
  • 08_token_counting.exs - Token counting, code vs prose comparison
  • 09_safety_settings.exs - Safety categories and thresholds

... (truncated)

Commits

• 42f97ae Release v0.8.6: Update default models to Gemini 2.5
• 4d3bbf5 Release v0.8.6: Update default models to Gemini 2.5 generation
• 3df6965 Release v0.8.5: Gemini 3 thinking levels, built-in tools, Veo 3.x inputs
• 2ee864f feat: Port 20 commits from Python SDK v1.56.0 updates
• a36c0ce Release v0.8.4: Add comprehensive examples suite and fix function calling
• 3e83cd6 Release v0.8.3: Interactions API parity with CRUD, streaming, and resumption
• f14a3e6 Release v0.8.2: Model path normalization, function response contract
• 6dc0816 Release v0.8.1: docs
• a1029ca Release v0.8.0: API parity with Python SDK
• 66e1b38 Release v0.7.3: System instructions, complete function calling framework, and...
• Additional commits viewable in compare view

Updates hammer from 7.0.1 to 7.1.0

Changelog

Sourced from hammer's changelog.

7.1.0 - 2025-07-18

• Fix key type inconsistency in backend implementations - all backends now accept term() keys instead of String.t() (#143)
• Add comprehensive test coverage for various key types (atoms, tuples, integers, lists, maps)
• Fix race conditions in atomic backend tests (FixWindow, LeakyBucket, TokenBucket)
• Replace timing-dependent tests with polling-based eventually helper for better CI reliability
• Add documentation warning about Redis backend string key requirement
• Fix typo in inc/3 optional callback documentation (#142)

Commits

• a57bdec improve changelog last commit (#145)
• bb061c5 Bump version to 7.1.0 (#144)
• 7d7967f Fix key type inconsistency in backend implementations (#143)
• 94d3952 Fixes typo for inc/3 optional callback @doc (#142)
• 79ca221 Bump benchee from 1.3.1 to 1.4.0 (#135)
• a09bbd0 Bump ex_doc from 0.37.3 to 0.38.2 (#141)
• d06a17b Bump credo from 1.7.11 to 1.7.12 (#134)
• 26df742 Update bug_report.md (#133)
• b8765fe Bump ex_doc from 0.37.2 to 0.37.3 (#131)
• See full diff in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot merge will merge this PR after your CI passes on it
• @dependabot squash and merge will squash and merge this PR after your CI passes on it
• @dependabot cancel merge will cancel a previously requested merge and block automerging
• @dependabot reopen will reopen this PR if it is closed
• @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore <dependency name> major version will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
• @dependabot ignore <dependency name> minor version will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
• @dependabot ignore <dependency name> will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
• @dependabot unignore <dependency name> will remove all of the ignore conditions of the specified dependency
• @dependabot unignore <dependency name> <ignore condition> will remove the ignore condition of the specified dependency and ignore conditions