fix(text): remove parentheses replacement in AttachmentToText

[Will-hxw]

Summary

Removes the unconditional replacement of ( with [ and ) with ] in AttachmentToText(). This was breaking markdown link format [text](url) by converting it to [text][url].

Problem

The AttachmentToText function was replacing all parentheses with square brackets:

result = strings.ReplaceAll(result, "(", "[")
result = strings.ReplaceAll(result, ")", "]")

This caused markdown links like [Read timeout](https://sentry.io/issues/123) to become [Read timeout][https://sentry.io/issues/123], which is not valid markdown.

Fix

Removed both replacement lines. The original CSV-safety escaping is unnecessary since the output format uses ; as delimiter, not comma.

Test plan

• Go build passes on cmd package
• Manual verification that markdown links remain intact in output

[Will-hxw]

Hi @140am @korotovsky @JAORMX , sorry to bother you. This PR has been open for a while, so I wanted to kindly check whether you might have time to review it or let me know if any changes are needed.I’m happy to make adjustments if required. Thanks!

[140am]

tested this a bit more against a real slack attachment payload and i think this is directionally right but still incomplete

the simple unit-level case [Read timeout](https://sentry.io/issues/123) is fixed by removing the paren replacement, but live slack returned the attachment title as:

[Read timeout](<https://sentry.io/issues/123>)

with the url wrapped in <...>

with the current PR, that still comes back through the MCP as:

Title: Read timeouthttps://sentry.io/issues/123

so i think the markdown parser also needs to handle the slack-shaped [text](<url>) form, something like:

regexp.MustCompile(`\[([^\]]+)\]\(<?(https?://[^)>]+)>?\)`)

i also tested the separate title + title_link attachment case, since #278 mentions TitleLink. that still comes back as just Title: Read timeout, because AttachmentToText does not currently read slack.Attachment.TitleLink

so the fix likely needs two pieces:

1. keep this paren replacement removal
2. support [text](<url>) in filterSpecialChars
3. optionally render TitleLink as [Title](TitleLink) before ProcessText

would also be good to add regression tests for both live-shaped cases

[Will-hxw]

Good catch on the `<...>` wrapper format! I've pushed an update that handles the angle-bracket markdown link style `text`.

Also added `TitleLink` support in `AttachmentToText` so `Title: text (url)` format is included.

Both changes are on `fix/278-parentheses-markdown-link` branch with test coverage. Let me know if there are other edge cases to cover.

[140am]

the [text](<url>) live slack case looks fixed now

i reran the same live attachment fixture through the updated PR and the markdown-title case now comes back cleanly:

Title: https://sentry.io/issues/123 - Read timeout, Text: markdown attachment fixture

one remaining issue: the separate title + title_link case still comes back with punctuation leakage:

Title: Read timeout https://sentry.io/issues/123); Text: title_link attachment fixture

looks like this is from formatting it as Title: %s (%s). changing that to markdown form should let the existing parser normalize it consistently:

parts = append(parts, fmt.Sprintf("Title: [%s](%s)", att.Title, att.TitleLink))

i tested that locally and against the live fixture, and it produced:

Title: https://sentry.io/issues/123 - Read timeout, Text: title_link attachment fixture

would also add a regression test for slack.Attachment{Title, TitleLink} since the current new tests cover [text](<url>) but not the TitleLink path