[experiment] Use unhygienic token matching for identifying macro variables

[petrochenkov]

All other comparisons in macro_rules use "stupid" token matching instead of name resolution.

macro m(a $a: ident) {
    $a /* <- NEW!: matches "$a" from the macro LHS regardless of hygienic contexts */
}

m!(a /* <- OLD: matches "a" from the macro LHS regardless of hygienic contexts */ b)

r? @ghost

[petrochenkov]

@bors try

[bors]

⌛ Trying commit 739bcfe with merge e6563ac70cbfd3196492061900e9e9793e065e4d...

[petrochenkov]

Inspired by #68746.

[bors]

☀️ Try build successful - checks-azure
Build commit: e6563ac70cbfd3196492061900e9e9793e065e4d (e6563ac70cbfd3196492061900e9e9793e065e4d)

[petrochenkov]

@craterbot check

[craterbot]

👌 Experiment pr-69410 created and queued.
🤖 Automatically detected try build e6563ac70cbfd3196492061900e9e9793e065e4d
🔍 You can check out the queue and this experiment's details.

ℹ️ Crater is a tool to run experiments across parts of the Rust ecosystem. Learn more

[craterbot]

🚧 Experiment pr-69410 is now running

ℹ️ Crater is a tool to run experiments across parts of the Rust ecosystem. Learn more

[craterbot]

🎉 Experiment pr-69410 is completed!
📊 3 regressed and 0 fixed (90999 total)
📰 Open the full report.

⚠️ If you notice any spurious failure please add them to the blacklist!
ℹ️ Crater is a tool to run experiments across parts of the Rust ecosystem. Learn more

[petrochenkov]

Regressions:

• rbalicki2.rustactoe.78e20cb04be08e6b7e56a72f3f6fe0f14a18dda4 - I couldn't reproduce this one.
• defmac-0.2.1 - the macro intentionally uses macro variables with same name and different contexts.
• ndarray-0.13.0 - depends on defmac-0.2.1 in --test mode.