Add findPrecedingToken

[gabritto]

The new code ports the original findPrecedingToken as faithfully as possible.

From the baselines, there are two differences:

1. The range of some returned tokens is different in Corsa.
2. In Strada, when the position is inside a JSDoc comment that has no tags, findPrecedingToken returns undefined. In Corsa, right now, we return the JSDoc. I am not sure if this difference matters or how to properly deal with that, so I thought I should leave this as a future work item.

[Copilot]

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

Comments suppressed due to low confidence (3)

internal/astnav/tokens_test.go:79

• [nitpick] Consider adding an inline comment to explain the purpose of the 'includeEOF' parameter, which would help clarify its intended effect in token retrieval.

func baselineTokens(t *testing.T, testName string, includeEOF bool, getTSTokens func(fileText string, positions []int) []*tokenInfo, getGoToken func(file *ast.SourceFile, pos int) *tokenInfo) {

internal/astnav/tokens_test.go:239

• [nitpick] Document the newly introduced 'position' parameter in writeRangeDiff to clarify how it influences the computation of token positions in the diff output.

func writeRangeDiff(output *strings.Builder, file *ast.SourceFile, diff tokenDiff, rng core.TextRange, position int) {

internal/astnav/tokens.go:395

• Instead of panicking when a whitespace-only token is encountered, consider handling this case gracefully (for example by returning an error or a fallback token) to avoid potential runtime crashes in production.

if result != nil && ast.IsWhitespaceOnlyJsxText(result) { panic("Expected result to be a non-whitespace token.") }

[andrewbranch]

Baseline diff looks good—I think those are Strada bugs (that probably never matter in practice).