feat: Update to claude-agent-sdk 0.2.119 (#587)

Bumps `@anthropic-ai/claude-agent-sdk` from `0.2.114` to `0.2.119`
(latest).

## Why

Primarily to pick up the April 20 system prompt verbosity revert called
out in the [April 23
postmortem](https://www.anthropic.com/engineering/april-23-postmortem),
which shipped in `claude-agent-sdk` `0.2.116` (parity with Claude Code
v2.1.116). Per the postmortem, an instruction added on April 16 —
_"Length limits: keep text between tool calls to ≤25 words. Keep final
responses to ≤100 words unless the task requires more detail."_ — caused
a measurable drop in coding quality for Sonnet 4.6, Opus 4.6, and Opus
4.7, and was reverted in the v2.1.116 release.

The other two regressions described in the postmortem (the
cache-clearing thinking-blocks bug fixed on April 10, and the default
reasoning effort revert on April 7) were already addressed before the
`0.2.114` this repo currently pins, so this upgrade is specifically
about the verbosity-prompt fix and the SDK improvements that have landed
since.

## What's also included between 0.2.114 and 0.2.119

From the [SDK release
notes](https://github.com/anthropics/claude-agent-sdk-typescript/releases):

- **0.2.116** — parity with Claude Code v2.1.116 (postmortem fix)
- **0.2.117** — parity with Claude Code v2.1.117
- **0.2.118** — adds `Options.managedSettings` for embedders to pass
policy-tier settings to the spawned CLI in-memory, honored below
IT-controlled managed sources
- **0.2.119**
- `excludeDynamicSections` now keeps static auto-memory instructions in
the cacheable system-prompt block; only the per-user memory directory
path and per-machine environment values are relocated to the first user
message
- Long-running SDK sessions now reconnect claude.ai-proxied MCP servers
after a transport-stream abort
- `SessionStore.append()` failures are now retried up to 3 times with
short backoff before the batch is dropped and `mirror_error` is emitted

No source changes were required in this repo — the public API surface
this project uses is unchanged.

## Verification

- `npm run build` ✅
- `npm run check` (lint + prettier) ✅
- `npm run test:run` ✅ (197 passed, 11 skipped)

---------

Co-authored-by: Ben Brandt <benjamin.j.brandt@gmail.com>

.github/dependabot.yml

@@ -8,7 +8,7 @@ updates:
   - package-ecosystem: "npm"
     directory: "/"
     schedule:
-      interval: "weekly"
+      interval: "daily"
     allow:
       - dependency-type: "all"
     groups: