Baixar agent-lsp 0.23.0 per a Windows

### Changed
• **`find_symbol` and `find_references` qualify empty results with an index-coverage caveat** ([#42](https://github.com/blackwell-systems/agent-lsp/issues/42))：some language servers index only opened documents (observed and wire-verified on `mql-lsp-server` v2.4.2 through 2.5.0-rc.5: `workspace/symbol` and `textDocument/references` return `[]` for symbols in files the session never `didOpen`ed, while `definition` resolves through live include-graph resolution), so an empty result can be an index-coverage artifact rather than a true "not found". `find_symbol` now distinguishes the empty causes: a server that does not declare `workspaceSymbolProvider` gets an explicit "workspace symbol search is unavailable" hint (previously the capability gate silently produced an empty result), and a server known to index only opened documents gets "this language server indexes only opened documents, so symbols or references in files not yet opened may be missing from this result; open the declaring file (open_document) and retry". `find_references` appends the same caveat to the qualified dead-code hint from #38/#46 when the workspace holds files the session never opened. The caveat is limited to servers verified to index only opened documents (currently `mql-lsp-server`, matched by executable or reported server name): whole-workspace indexers such as gopls, rust-analyzer, and pyright answer for unopened files, so their empty results stay authoritative and unhedged. For those servers, a bounded workspace probe (incremental directory reads capped at 2000 total entries — files, directories, and excluded entries all count toward the budget; budget exhaustion, walk errors, and cancellation report unknown coverage and still show the caveat) compares the workspace's files in that server's languages (other files cannot be missing from its index) against the session's opened path set, so a fully-opened workspace keeps the plain hint and opens outside the eligible set (e.g. `node_modules`) cannot mask unopened source files. Empty `find_symbol` responses keep the established response envelope for every detail level — the caveat is appended to the detail-specific encoding (the `workspaceSymbolsResponse` object with `total`/`symbols` for hover detail) rather than replacing it with a bare list. `go_to_definition` is deliberately untouched — its empty result is authoritative (position resolved to nothing) and the fuzzy-fallback provenance rules from #40 apply there. Cache-served provenance (`refs.db` results that the underlying server cannot produce live) remains future work tracked in #42.

### Fixed
• **`restart_lsp_server` no longer requires `root_dir` and re-opens the documents that were open** ([#54](https://github.com/blackwell-systems/agent-lsp/issues/54))：the schema marked `root_dir` optional but the handler rejected calls without it; it now defaults to the current workspace root. After a restart the fresh server previously knew no open documents, so servers that index only opened files (such as `mql-lsp-server`) returned empty workspace symbols until a document was reopened. The documents open before the restart are now re-sent with `didOpen`, using their current content read from disk through the workspace-root confinement checks, including when the workspace root is reached through a symlink. Contributed by [@davalillo](https://github.com/davalillo).
• **Reading from a language server could spend minutes of CPU on output with no message header** ([#65](https://github.com/blackwell-systems/agent-lsp/issues/65))：the LSP frame reader searched the whole accumulated buffer for the header terminator on every read, which is quadratic when a server writes a large amount of output without a `Content-Length` header (10 MB arriving in small reads took about three minutes). The search now resumes where the last one stopped. Normal traffic was not affected, since a message header sits at the start of the buffer. As a side effect the `internal/lsp` test suite runs in seconds instead of minutes, and its `-race` run no longer hits the timeout.
• **`blast_radius` filters parameters and locals on flat `documentSymbol` servers and names real callers** ([#53](https://github.com/blackwell-systems/agent-lsp/issues/53))：servers that publish a flat symbol list (`mql-lsp-server` reports parameters and locals as top-level Function/Variable entries) bypassed the #49 nested-symbol filter, so parameters and locals still became blast-radius targets, and every caller was labeled with the queried symbol's own name, producing degenerate self-edges. Flat lists are now re-nested by range, re-nested parameter/local shapes under a function are excluded as targets and as caller candidates, and each caller is named after the function that contains the reference, with deduplicated caller-to-target edges. Member functions under a class or namespace stay targets, symbols nested through `containerName` get the same filtering, a changed symbol that calls another changed symbol stays a target, and a reference inside a TS/JS arrow-function constant is attributed to that constant. Contributed by [@davalillo](https://github.com/davalillo).
• **`safe_apply_edit` previewed the wrong span on lines with non-ASCII text**：it located `old_text` by byte offset and passed byte counts as LSP columns. That happened to agree with the simulation while the edit path also counted bytes; once [#52](https://github.com/blackwell-systems/agent-lsp/issues/52) made that path count UTF-16 units (0.22.0), a match preceded by non-ASCII text on its line was previewed at a shifted span (for example `x := "café"; y := 2` with `y := 2` replaced previewed as `...; yy := 3`), so the diagnostic gate judged an edit that was not the one requested. Files written to disk were not affected. It now shares one byte-to-UTF-16 conversion with `apply_edit`. `safe_delete_symbol` and `blast_radius` had the same byte-column mistake when locating a symbol on its declaration line, which could make the reference lookup miss when non-ASCII text preceded the name; they now convert too.