feat: A better logic for detecting the thinking prefix in the response. (#423)

* fix: Check response prefix.

In case if the model adds <think> at the end of user prompt and only generates </think> end, then the detection goes wrong.

* fix: Handle whitespace in CoT, remove redundant checks

* fix: Type checker

* fix: try looking at tag positions, not text end

regex

* ruff, fix import sorting

* fix: rich markup

* fix: I missed other ones

* feat: Update SHA256SUMS file hashes in the tests.

A major change that affects reproducibility.

* fix: It is now sensible to also update the extra SHA256SUMS.ci2 file.

* fix: Only consider whitespace, no other text or instructions.

because mistral-3 as additional reasoning instructions in its chat template. And I suppose many other models can have it too.

* fix: Update windows hashes.

* fix: Update CI hashes too

* as always update case two of mistral-3 (ci2 hash)

* docs: update comment

* feat: Handle the edge case for models having additional instructions.

* fix: Update windows hash for mistral-3

* docs: remove a line from the comments

because I'm not sure about GPT-OSS models' thinking tags and it cannot be confirmed using an untrained tiny GPT-OSS model. And inference fallback would of course generate gibberish as the model cannot understand additional instructions about 'how to generate response and how to think' from the chat_template.

* fix: a few things.

* fix: Update hash for qwen3.5 after the whitespace fix for its response prefix.

* docs: Update comment

* fix: Update qwen3.5 hash for CI

* fix: Remove Case 2 which only serves tests

unnecessary

* fix: Hash

* fix: concern is valid enough, so we use a small text.

add a comment too

* docs: minor

tests/mistral-3/SHA256SUMS.windows

@@ -1,7 +1,7 @@
 72f84af4ea36b82409c35e31b584361534305ef7c0d90fce20d0dc38a7efead8 *chat_template.jinja
 e4c5278b361c57621253c27a2c3db358e1580aec8a14be8e19d4420a224137cf *config.json
 8dde85c000ae807be907421465826c7c63a39f6acf6d04a5a84efaf116ed4ef7 *generation_config.json
-29aff97d5633dead9e1ccd29a2cc153b4b7431d22f63c8d6cf60bc6547681cc9 *model.safetensors
+20b5a820b38438202c64e4fc9807bd19e29678bebd678d29b2ee2d2f5bf71587 *model.safetensors
 20e7a6dcde0a6f60ea3b4fb08f6f7afa62532dda93a3111e28384ba5150575f9 *processor_config.json
 c3a8d92e371b92a2cd6e678e31ebc27d0235e929a51fbf290f74742b341fa96f *tokenizer.json
 60a8042e29b4b20e884e48375aa1b9ac0025547371d50e60f6d55e6a9675e868 *tokenizer_config.json

tests/qwen3.5-moe/SHA256SUMS.ci

@@ -1,7 +1,7 @@
 a4aee8afcf2e0711942cf848899be66016f8d14a889ff9ede07bca099c28f715 *chat_template.jinja
 749b56d1b1e08081981169db6f2c44ab0be4fd6ebb452d15baafa5e09c21586a *config.json
 4625d1d64d41d1fa9dae7af4ba1e1d7e65a194073d4efa58acb266a916eaaa74 *generation_config.json
-5fb94c65bcd9d736735a45e50c2b0bfafd3bb09a444c49b8cff2e131ed35797e *model.safetensors
+2b3e575ac065f11ae5d4a7c3740efccbed294b646f1645239191ee8393354e03 *model.safetensors
 01562eddd6f9e9ec4bc31656a3b7055284cafbf889acc6c4348dca431ae31f68 *processor_config.json
 87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4 *tokenizer.json
 2e31d1126e81bddf8d15c3f95260fb487b48c5131b24fcbb5bb9d2537e7afac0 *tokenizer_config.json

tests/qwen3.5-moe/SHA256SUMS.windows

@@ -1,7 +1,7 @@
 a92e1dd97cb1cb175c9b70c0828e146bea4371c2643319b661b777e89811972e *chat_template.jinja
 b75e911805663da79fb9fbbbcc917b8f1a285d2da54d95c2c63ea7c1ffe9a05a *config.json
 2cbd9df0e99570efcced23b8d777bdf1fc692efda54b21eb59ad56ade76c9db6 *generation_config.json
-5f099b32807d0b84ed90765ca0ed53f8771da4738767bc1940486fec954570cf *model.safetensors
+6061519a9595326df41abcdd093892463793d4d026d6fd23548f1792f622a252 *model.safetensors
 0c29f9491e769aabbc389ad5912127cf6d9d5fceda2db8767f73d48131348c81 *processor_config.json
 87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4 *tokenizer.json
 4796e48d790a26d65f167bec8fc742beaa71f79f9468a6cd8b3ffa97f6e2a198 *tokenizer_config.json